Jou nal o
Clinical Medicine
A icle
Fac o s Associa ed wi h he P esc ibing o
High-In ensi y S a ins
A mando Chau e-Pa dos 1,2,3, Sa a Malo 1,2,3,* , Ma ía JoséRabanaque 1,2,3 ,
Fede ico A ibas 1,2,3, Belén Mo eno-F anco 1,2 and Isabel Aguila -Palacio 1,2,3
1Depa men o P e en i e Medicine and Public Heal h, Uni e si y o Za agoza, 50009 Za agoza, Spain;
[email p o ec ed] (A.C.-P.); abanake@uniza .es (M.J.R.); [email p o ec ed] (F.A.);
[email p o ec ed].es (B.M.-F.); iaguila @uniza .es (I.A.-P.)
2Fundación Ins i u o de In es igación Sani a ia de A agón (IIS A agón), 50009 Za agoza, Spain
3GRISSA Resea ch G oup, 50009 Za agoza, Spain
*Co espondence: smalo@uniza .es; Tel.: +34-976-761-761
Recei ed: 19 Oc obe 2020; Accep ed: 25 No embe 2020; Published: 27 No embe 2020


Abs ac :
In his s udy, we in es iga ed he ela ionship be ween sociodemog aphic, clinical,
an h opome ic, and li es yle cha ac e is ics and he ype o s a in p esc ibed o p ima y p e en ion
o ca dio ascula disease (CVD). We conduc ed an obse a ional s udy in wo ke s who began
s a in ea men . S a in he apy was ca ego ized as “high-in ensi y” o “low–mode a e-in ensi y”.
Wo ke s we e classi ied acco ding o he alignmen o hei s a in he apy wi h he ecommended
managemen p ac ices. Logis ic eg ession models we e used o e alua e he associa ion be ween he
di e en a iables s udied and he p obabili y o being p esc ibed high-in ensi y s a ins. The only
a iables associa ed wi h a highe p obabili y o being ea ed wi h high-in ensi y s a ins we e
inc eased physical ac i i y (>40 e sus <20 METs (me abolic equi alen o ask) h/wk; odds a io
(OR), 1.65; 95%CI, 1.08–2.50) and, in diabe ics, highe low-densi y lipop o ein choles e ol (LDL-C)
le els (
≥
155 mg/dL e sus <155 mg/dL; OR, 4.96; 95%CI, 1.29–19.10). The model ha bes p edic ed
ea men in ensi y included LDL-C, diabe es, hype ension, smoking, and age (a ea unde he
Recei e Ope a ing Cha ac e is ic cu e (AUC), 0.620; 95%CI, 0.574–0.666). The p esc ibing and
ype o s a in used in p ima y CVD p e en ion did no co espond wi h he indica ions in cu en
guidelines. The p obabili y o ecei ing high-in ensi y s a ins was highe in diabe ics wi h high LDL-C
le els and in mo e physically ac i e indi iduals. These indings unde sco e he g ea a iabili y and
unce ain y in he p esc ibing o s a ins.
Keywo ds:
ca dio ascula diseases; p e en ion and con ol; heal hy li es yle; hyd oxyme hylglu a yl-
coa educ ase inhibi o s; p ac ice pa e ns; physicians’
1. In oduc ion
Clinical p ac ice guidelines o p e en ca dio ascula disease (CVD) a e based on isk assessmen ,
ecommenda ion o a heal hy li es yle, and, in some cases, pha macological ea men , including lipid-
lowe ing he apy [
1
,
2
]. Cu en Eu opean guidelines o CVD p e en ion conside s a ins as a
i s -line lipid-lowe ing op ion [
1
]. These d ugs ha e p o en e icacy in he p ima y p e en ion o
CVD, educing mo bidi y and mo ali y in indi iduals wi h mode a e and high isk [
3
]. Howe e ,
hei e ec i eness in low- isk indi iduals emains a opic o deba e [4,5].
Acco ding o hei abili y o lowe blood le els o low-densi y lipop o ein choles e ol (LDL-C),
s a ins a e ypically classi ied as low-, mode a e-, o high-in ensi y [
6
,
7
]. This classi ica ion depends on
he d ug and he dose used. Guidelines do no indica e he ype o s a in ha should be p esc ibed,
bu es ablish a ge LDL-C le els ha depend on he isk o CVD. Thus, when deciding he ype o
J. Clin. Med. 2020,9, 3850; doi:10.3390/jcm9123850 www.mdpi.com/jou nal/jcm
J. Clin. Med. 2020,9, 3850 2 o 12
s a in o be p esc ibed, he physician mus be awa e o he pa ien ’s CVD isk, as well as cu en and
a ge LDL-C le els.
Gi en hese a he nonspeci ic ecommenda ions, i is o in e es o know which ac o s (including
classical CVD isk ac o s, basal CVD isk, and li es yle- ela ed ac o s) a e conside ed mos impo an
in clinical p ac ice when p esc ibing s a ins. This knowledge can be use ul o guide p esc ip ion and
o adjus i o es ablished s anda ds. Some s udies ha e desc ibed he cha ac e is ics o s a in use s
wi hou s a i ying by s a in in ensi y [
8
,
9
]. O he s ha ha e collec ed in o ma ion om adminis a i e
da abases bene i om la ge sample sizes, bu end o include incomple e clinical da a and lack
in o ma ion on pa ien li es yles [
10
]. Finally, some au ho s ha e ocused hei analyses on indi iduals
al eady in ea men , neglec ing new use s and he e o e no conside ing he cha ac e is ics o he use
a he momen o he decision o ea [11].
The aim o he p esen s udy was o cha ac e ize he p o ile o new use s o s a ins o p ima y
CVD p e en ion in a coho o wo ke s, and o in es iga e he associa ion be ween hese cha ac e is ics
and he ype o s a in p esc ibed.
2. Expe imen al Sec ion
2.1. S udy Design and Pa icipan s
This s udy was based on he A agon Wo ke ’s Heal h S udy coho (AWHS). The AWHS is a
longi udinal p ospec i e coho s udy ha was designed o e alua e he e olu ion o adi ional and
eme gen CVD isk ac o s and hei associa ion wi h he p e alence and p og ession o subclinical
a he oscle osis in a popula ion o middle-aged ca plan wo ke s in Spain [
12
]. Rec ui men was
ca ied ou o he mos pa in 2009 and 2010, and con inues oday. F om his popula ion, a subg oup
o 2667 wo ke s who we e aged 40–55 yea s a baseline was selec ed. This subg oup comple ed
addi ional ques ionnai es abou ca dio ascula and li es yle ac o s, including die , physical ac i i y,
and sleep habi s. The wo ke s we e also in i ed o unde go a nonin asi e imaging examina ion o
subclinical a he oscle osis.
Fo he p esen s udy, we selec ed om he a o emen ioned subg oup wo ke s who began s a in
he apy o p ima y p e en ion o CVD be ween 1 July 2010 and 31 Decembe 2018. New use s
we e de ined as hose who did no ecei e a p esc ip ion o s a ins du ing he p eceding 6 mon hs.
Wo ke s in ea men o p ima y p e en ion o CVD we e hose wi h no eco ded hospi al admission
o CVD in he hospi aliza ion da abase o he A agon Heal h Sys em p eceding hei i s s a in
p esc ip ion. Pa ien s wi h CVD we e de ined as hose wi h a main diagnosis co esponding o any o
he ollowing In e na ional Classi ica ion o Diseases 10 h e ision (ICD-10) codes: G45, G46, G81–G83,
I20–I28, I46, I49.0, I50, and I60–I79.
2.2. Da a Sou ces and Va iables
In o ma ion on s a in ea men was ob ained om he Fa masalud da abase, which collec s
in o ma ion on d ugs dispensed by pha macies and unded by he A agon Heal h Sys em.
D ugs we e classi ied acco ding o he 2019 ATC/DDD (Ana omical The apeu ic Chemical Classi ica ion
Sys em/De ined Daily Dose) sys em, de ined by he Wo ld Heal h O ganiza ion. We iden i ied wo ke s
wi h a leas one p esc ip ion du ing he s udy pe iod o a d ug co esponding o he ollowing ATC
codes: C10AA (hyd oxyme hylglu a yl-CoA (HMG CoA) educ ase inhibi o s), C10BA (HMG CoA
educ ase inhibi o s in combina ion wi h o he lipid-modi ying agen s), C10BX (HMG CoA educ ase
inhibi o s, o he combina ions).
In o ma ion on o he CVD ea men s was also collec ed. Agen s co esponding o he ollowing ATC
codes we e classi ied as an ihype ensi es: C02 (an ihype ensi es), C03 (diu e ics), C07 (be a-blocking
agen s), C08 (calcium channel blocke s), and C09 (agen s ac ing on he enin–angio ensin sys em).
P esc ip ions o he ea men o diabe es we e de ined as hose co esponding o ATC code A10 (d ugs
used in diabe es). The ollowing c i e ion was applied o de ine indi iduals ecei ing co- ea men s:
J. Clin. Med. 2020,9, 3850 3 o 12
he exis ence o a leas one d ug p esc ip ion co esponding o he a o emen ioned codes du ing he
yea p eceding he i s s a in p esc ip ion. Gi en ha diu e ics and be a-blocking agen s a e d ugs also
p esc ibed o o he indica ions, p esc ibing o hese d ugs was only conside ed co- ea men i he pa ien
ecei ed a leas h ee dis inc p esc ip ions wi hin he same yea . We conside ed ha his equency was
indica i e o egula use. E en so, o assess he possible bias o his assump ion, we pe o med sensi i i y
analyses: (i) conside ing hem as co- ea men i he pa ien ecei ed a leas one p esc ip ion in he same
yea ; (ii) no conside ing hem as co- ea men a all.
In addi ion o in o ma ion on o he CVD p e en i e ea men s, o all pa icipan s in he
p esen s udy, we ex ac ed in o ma ion on wo king cha ac e is ics, clinical and an h opome ic
measu emen s, smoking s a us, and diagnoses o o he diseases ha may modi y CVD isk. Fo hese
pa ame e s, we selec ed o each wo ke da a collec ed as close as possible be o e he da e o he i s
s a in p esc ip ion.
Da a pe aining o each indi idual wo k si ua ion included he numbe o yea s wo king in he
ac o y, wo k ype, and wo k shi . Wo k ype was classi ied as manual o seden a y. Wo k shi s in
he plan we e classi ied as o a ing mo ning/a e noon shi , o a ing mo ning/a e noon/nigh shi ,
cen al shi , and nigh shi .
Fo clinicalandan h opome iccha ac e iza ion, weex ac ed he ollowingda a, whichwe e eco ded
in annual medical exams as desc ibed in Casasno as e al. [
12
]: heigh , weigh , se um c ea inine, o al and
high-densi y lipop o ein (HDL)-choles e ol, iglyce ides, se um glucose, whole blood glycosyla ed
hemoglobin (HbA1c), and blood p essu e. Annual da a on smoking s a us we e also ex ac ed.
Diseases conside ed o modi y he isk o CVD we e heuma oid a h i is (RA), a hy hmias,
and ansien ischemic acciden . Diagnoses co esponding o hese diseases (In e na ional Classi ica ion
o P ima y Ca e codes L88, K80, and K89, espec i ely) we e ex ac ed om he A agon p ima y
ca e da abase.
Li es yle o he s udy pa icipan s was cha ac e ized based on alcohol in ake, adhe ence o a
Medi e anean die , physical ac i i y, sleep, and seden a y ime. A semiquan i a i e ood- equency
ques ionnai e p e iously alida ed in Spain [
13
] was used o assess die a y habi s. Leisu e ime physical
ac i i y was assessed wi h he alida ed Spanish e sion [
14
] o he physical ac i i y ques ionnai e
used in he Nu ses’ Heal h S udy and Heal h P o essionals’ Follow-up S udy. To es ima e seden a y
ime, we included he numbe o hou s ( om “ne e ” o “9 o mo e han 9 h a day”) spen si ing,
as epo ed by he pa icipan , du ing bo h wo king and leisu e ime in a ypical weekday. The same
p ocedu e was ollowed o collec in o ma ion on he numbe o hou s spen sleeping pe nigh .
The p esence o subclinical a he oscle osis in bo h ca o id and emo al a e ies was de e mined by
ul asound, acco ding o he p ocedu es desc ibed in Laclaus a e al. [
15
]. Plaque was de ined as a ocal
s uc u e p o uding
≥
0.5 mm in o he lumen o eaching a hickness
≥
50% o he su ounding in ima.
2.3. Analyses
S a in he apy was ca ego ized as “high-in ensi y” (a o as a in o osu as a in) o “low–mode a e-
in ensi y” (sim as a in, lo as a in, p a as a in, lu as a in, o pi a as a in) [7].
Clinical and an h opome ic a iables we e de ined as ollows. Glome ula il a ion a e
(eGFR) was es ima ed using he modi ica ion o die in enal disease (MDRD) o mula [
16
], and an
eGFR <60 mL/min/1.73 m2
was conside ed indica i e o ch onic kidney disease [
17
]. Low-densi y
lipop o ein choles e ol (LDL-c) was es ima ed using F iedewald’s es ima ion [
18
]. Applying he isk
ac o de ini ions p oposed by he 2016 Eu opean guidelines on CVD p e en ion in clinical p ac ice [
1
],
we de ined a wo ke as diabe ic i he was ea ed wi h an idiabe ics, had a HbA1c
≥
6.5%, o had as ing
plasma glucose
≥
126 mg/dL. C i e ia o hype ension we e sys olic blood p essu e
≥
140 mmHg,
dias olic blood p essu e
≥
90 mmHg, o exis ence o a p esc ip ion o an ihype ensi e ea men .
To assess adhe ence o a Medi e anean die , we used he Al e na e Medi e anean Die a y Index
(aMED) [
19
]. To de e mine he o al physical ac i i y pe o med by each wo ke , METs (me abolic
equi alen o ask) we e assigned o each ac i i y [
20
] and mul iplied by he numbe o hou s pe
J. Clin. Med. 2020,9, 3850 4 o 12
week ha he wo ke dedica ed o he ac i i y. The o al amoun o MET-h/week was calcula ed as he
sum o he MET-h o he di e en ac i i ies. We used 5.5 h/day as a cu -o poin o seden a y ime,
ollowing p e ious indings o inc eased body mass index (BMI), wais ci cum e ence, and insulin
le els in his same coho [
21
]. Acco ding o sleeping ime, wo ke s we e di ided in o wo g oups:
<6 h/nigh
o
≥
6 h/nigh . This decision was based on he lowe limi in he Na ional Sleep Founda ion’s
ecommenda ions o adul s [22].
Finally, we calcula ed he 10 yea isk o a i s a al a he oscle o ic e en using he Sys emic
Co ona y Risk Es ima ion (SCORE) o low- isk coun ies [
23
]. Nex , pa icipan s we e classi ied
acco ding o Eu opean guidelines [
1
] as low, mode a e, high, o e y high isk by combining hei
SCORE and o he CVD isk ac o s including diabe es, ch onic kidney disease, and ex eme LDL-C o
blood p essu e alues.
Based on his isk classi ica ion and blood LDL-C le els, Eu opean guidelines ecommend
ce ain a ge LDL-C le els ha should be achie ed o educe CVD isk. Following hese guidelines,
we classi ied wo ke s acco ding o he ecommended managemen p ac ices as ollows: “s a ins no
indica ed” i , based on CVD isk and LDL-C le els, only li es yle ad ice is ecommended; “LDL-C
educ ion <50%” i d ug ea men is indica ed and LDL-C should be educed by less han 50% o
each he a ge LDL-C le els; “LDL-C educ ion
≥
50%” i d ug ea men is indica ed and LDL-C
should be educed by a leas 50% o each a ge le els.
All a iables we e desc ibed o he global popula ion and s a i ied acco ding o s a in in ensi y.
Fo each a iable, we pe o med s a is ical analyses o assess he di e ence be ween ea men in ensi y
g oups. The S uden ’s - es was used o quan i a i e a iables. The chi-squa ed es was used o
analyze ca ego ical a iables when he numbe o wo ke s exceeded 30. Low- equency a iables we e
analyzed using Fishe ’s exac es .
To e alua e he associa ion be ween he di e en ac o s s udied (sociodemog aphic, clinical,
and li es yle) and he p obabili y o being p esc ibed high-in ensi y e sus low–mode a e-in ensi y s a ins,
we used logis ic eg ession models, wi h s a in in ensi y as a dependen a iable. Fi s , we compu ed
unadjus ed models. Nex , we de eloped mul i a iable models ha included adi ional CVD isk ac o s
and a iables o which we ob ained a p- alue <0.1 in he unadjus ed analysis. We also included
in e ac ion e ms in he model. Fo inclusion in he models, con inuous a iables we e ca ego ized using
p e-es ablished cu -o poin s. Since we did no know how hese a iables would beha e in he model,
we es ablished as a e e ence he ca ego y ha allowed us o ind signi ican di e ences be ween g oups
o , ailing ha , he lowes ca ego y.
Finally, o assess he abili y o CVD isk ac o s o p edic he p esc ibing o high-in ensi y s a ins,
we de eloped logis ic eg ession models. Explana o y a iables we e inco po a ed sequen ially
acco ding o he magni ude o hei associa ion in he p e ious mul i a iable models. P edic i e powe
was e alua ed by calcula ing he a ea unde he ecei e ope a ing cha ac e is ic (ROC) cu e (AUC).
All analyses we e pe o med using STATA e sion 14 (S a aCo p, College S a ion, TX, USA).
2.4. E hical Issues
All subjec s ga e hei in o med consen o inclusion be o e hey pa icipa ed in he s udy.
The s udy was conduc ed in acco dance wi h he Decla a ion o Helsinki, and he p o ocol was
app o ed by he E hics Commi ee o A agon (P ojec iden i ica ion code PI17/00042).
3. Resul s
A o al o 683 wo ke s began ea men wi h s a ins du ing he ollow-up. Table 1shows hei
sociodemog aphic, clinical, and an h opome ic cha ac e is ics. The mean age o new s a in use s was
53 yea s, 5% we e women, and mos pe o med manual wo k wi h o a ing shi s. Mo e han hal
we e classi ied as ha ing hype ension, 81% we e o e weigh o obese, and 10% had diabe es. Table 2
shows li es yle and imaging cha ac e is ics. Mo e han a hi d o he new s a in use s smoked a he
ime o p esc ip ion, 15% consumed mo e han 40 g o alcohol pe day, and 36% had low adhe ence o
J. Clin. Med. 2020,9, 3850 5 o 12
a Medi e anean die . Mo e han hal sa o a leas 5.5 h a day, and 20% slep less han 6 h pe nigh .
Subclinical a he oscle osis was mo e equen in he emo al a e y han in he ca o id a e y.
Table 1.
Sociodemog aphic, clinical, and an h opome ic cha ac e is ics o wo ke s acco ding o he
in ensi y o s a in p esc ibed.
S a in In ensi y
All (N =683) Low–Mode a e-In ensi y
(N =360)
High-In ensi y
(N =323) p
Sociodemog aphic
Mean (SD) age, y 53.0 (4.0) 52.8 (4.3) 53.1 (3.7) 0.390
Women, N (%) 31 (4.5) 18 (5.0) 13 (4.0) 0.585
Mean (SD) no. o yea s in ac o y 28.4 (4.5) 28.4 (4.9) 28.4 (4.1) 0.963
Wo k shi , N (%) 0.137
Ro a ing 522 (76.5) 283 (78.8) 239 (74.0)
Fixed 160 (23.4) 76 (21.2) 84 (26.0)
Wo k ype, N (%) 0.117
Assembly line/manual 590 (86.4) 318 (88.3) 272 (84.2)
Seden a y 93 (13.6) 42 (11.7) 51 (15.8)
Clinical and an h opome ic
BMI (kg/m2), N (%) 0.600
<25 127 (18.6) 67 (18.6) 60 (18.6)
25–30 392 (57.4) 212 (58.9) 180 (55.7)
≥30 164 (24.0) 81 (22.5) 83 (25.7)
EGFR <60 mL/min/1.73 m2, N (%) 8 (1.2) 6 (1.7) 2 (0.6) 0.292
To al choles e ol (mg/dL), mean (SD) 246.9 (37.5) 244.8 (35.0) 249.2 (40.0) 0.124
LDL-choles e ol (mg/dL), mean (SD) 159.0 (33.2) 157.7 (30.5) 160.6 (36.0) 0.258
HDL-choles e ol (mg/dL), mean (SD) 52.2 (11.5) 52.2 (11.4) 52.2 (11.6) 0.972
T iglyce ides (mg/dL), mean (SD) 187.8 (139.7) 185.1 (148.5) 190.7 (129.4) 0.601
Diabe es, N (%) 69 (10.1) 33 (9.2) 36 (11.2) 0.392
Hype ension, N (%) 354 (51.8) 175 (48.6) 179 (55.4) 0.075
Rheuma oid a h i is, N (%) 12 (1.8) 5 (1.4) 7 (2.2) 0.563
P e ious TIA, N (%) 3 (0.4) 2 (0.6) 1 (0.3) 1.000
A hy hmia, N (%) 15 (2.2) 7 (1.9) 8 (2.5) 0.795
Abb e ia ions: BMI, body mass index; EGFR, es ima ed glome ula il a ion a e; HDL, high-densi y lipop o ein;
LDL, low-densi y lipop o ein; N, numbe ; SD, s anda d de ia ion; TIA, ansien ischemic a ack. p- alues
co espond o - es /chi-squa ed es o Fishe ’s exac es .
Table 2. Li es yle cha ac e is ics and subclinical a he oscle osis among wo ke s, s a i ied by s a in in ensi y.
S a in In ensi y
All (N =683) Low–Mode a e-In ensi y
(N =360)
High-In ensi y
(N =323) p
Li es yle
Smoking, N (%) 0.678
Ne e 131 (19.2) 73 (20.3) 58 (18.0)
Cu en 249 (36.5) 127 (35.3) 122 (37.8)
Fo me 302 (44.3) 160 (44.4) 142 (44.0)

J. Clin. Med. 2020,9, 3850 6 o 12
Table 2. Con .
S a in In ensi y
All (N =683) Low–Mode a e-In ensi y
(N =360)
High-In ensi y
(N =323) p
Alcohol in ake, N (%) 0.685
Low isk (<40 g/d) 581 (85.1) 306 (85.0) 275 (85.1)
Medium isk (40–60 g/d) 64 (9.4) 36 (10.0) 28 (8.7)
High isk (>60 g/d) 38 (5.6) 18 (5.0) 20 (6.2)
Medi e anean die adhe ence, N (%) 0.379
Low 244 (35.7) 131 (36.4) 113 (35.0)
Medium 372 (54.5) 189 (52.5) 183 (56.7)
High 67 (9.8) 40 (11.1) 27 (8.4)
METs h/wk, mean (SD) 31.0 (21.2) 29.6 (20.4) 32.6 (22.0) 0.065
Si ing 5.5 h/day o mo e, N (%) 361 (53.1) 189 (52.5) 172 (53.8) 0.744
Sleep <6h/nigh , N (%) 136 (20.0) 64 (17.8) 72 (22.4) 0.130
Ca o id a he oscle osis, N (%) 273 (44.0) 130 (40.1) 143 (48.3) 0.040
Femo al a he oscle osis, N (%) 404 (65.6) 204 (61.1) 200 (70.9) 0.010
Abb e ia ions: METs h/wk, Me abolic equi alen s o ask hou s pe week; N, numbe ; SD, s anda d de ia ion.
p- alues co espond o - es /chi-squa ed es . Medi e anean die adhe ence: Al e na e Medi e anean Die a y
Index (aMED) sco e. Low: 0–3 sco e in aMed Index. Medium: 4–6 sco e in aMed Index. High: 7–9 sco e in aMed
Index. S a is ically signi ican di e ences (p<0.05) a e shown in bold.
High-in ensi y s a ins we e p esc ibed o 323 (47%) o he new s a in use s. Wo ke s p esc ibed
low–mode a e-in ensi y and high-in ensi y s a ins had simila cha ac e is ics. Mean choles e ol and
LDL-C le els we e highe in wo ke s p esc ibed high-in ensi y s a ins, al hough hese di e ences did
no each s a is ical signi icance. Subclinical a he oscle osis was mo e equen in wo ke s p esc ibed
high-in ensi y s a ins.
Figu e 1shows he dis ibu ion o wo ke s acco ding o hei CVD isk, managemen
(as ecommended by Eu opean guidelines), and s a in ea men in ensi y. The la ges g oup is
ha o wo ke s wi h a mode a e isk o CVD o whom pha macological ea men is ecommended
o educe LDL-C by less han 50%. In his g oup o 413 wo ke s, 220 (53%) we e p esc ibed
low–mode a e-in ensi y s a ins. The second-la ges g oup is ha o wo ke s wi h a high isk o CVD
o whom pha macological ea men is ecommended o educe LDL-C by mo e han 50%. In his
g oup o 94 wo ke s, 42 (45%) we e p esc ibed low–mode a e-in ensi y s a ins. A o al o 108 wo ke s
wi h low o mode a e isk did no mee he c i e ia o pha macological ea men . O hese, 45 (42%)
we e p esc ibed high-in ensi y s a ins.
Table 3shows he esul s o unadjus ed and mul i a iable logis ic models o e alua e he associa ion
be ween ca dio ascula isk ac o s and he p esc ibing o high- e sus low–mode a e-in ensi y s a ins.
Gi en ha we obse ed an in e ac ion be ween LDL-C le els and diabe es in he con ex o p esc ibing o
high-in ensi y s a ins, his in e ac ion e m was included in he models. Highe le els o physical ac i i y
(>40 e sus <20 METs h/wk; OR, 1.65; 95%CI, 1.08–2.50) and, in diabe ics, LDL-C le els
≥155 mg/dL
e sus <155 mg/dL (OR, 4.96; 95%CI, 1.29–19.10) we e associa ed wi h a highe p obabili y o
high-in ensi y s a in p esc ibing. In non-diabe ics, we obse ed no associa ion be ween he p esc ibing
o high-in ensi y s a ins and LDL-C le els. Simila ly, he e was no associa ion de ec ed be ween
he p esc ibing o high-in ensi y s a ins and age, smoking s a us, o he p esence o hype ension
o diabe es.
J. Clin. Med. 2020,9, 3850 7 o 12
J. Clin. Med. 2020, 9, x FOR PEER REVIEW 7 o 13
Figu e 1. Numbe o wo ke s p esc ibed low–mode a e- o high-in ensi y s a ins, s a i ied by
ca dio ascula disease (CVD) isk and ecommended managemen . CVD isk: classi ica ion o
ca dio ascula isk acco ding o cu en Eu opean guidelines on CVD p e en ion, based on he
combina ion o SCORE alue and he p esence o o he CVD isk ac o s. Recommended
managemen : ecommended pa ien managemen acco ding o cu en Eu opean guidelines on CVD
p e en ion, based on CVD isk, cu en LDL-choles e ol (LDL-C) le els, and a ge LDL-C le els. NI:
d ug ea men “no indica ed”, only li es yle ad ice ecommended. <50%: d ug ea men indica ed
o educe LDL-C by <50% o each a ge LDL-C le els. ≥50%: d ug ea men indica ed o educe
LDL-C by ≥50% o each a ge LDL-C le els.
Table 3 shows he esul s o unadjus ed and mul i a iable logis ic models o e alua e he
associa ion be ween ca dio ascula isk ac o s and he p esc ibing o high- e sus low–mode a e-
in ensi y s a ins. Gi en ha we obse ed an in e ac ion be ween LDL-C le els and diabe es in he
con ex o p esc ibing o high-in ensi y s a ins, his in e ac ion e m was included in he models.
Highe le els o physical ac i i y (>40 e sus <20 METs h/wk; OR, 1.65; 95%CI, 1.08–2.50) and, in
diabe ics, LDL-C le els ≥ 155 mg/dL e sus < 155 mg/dL (OR, 4.96; 95%CI, 1.29–19.10) we e associa ed
wi h a highe p obabili y o high-in ensi y s a in p esc ibing. In non-diabe ics, we obse ed no
associa ion be ween he p esc ibing o high-in ensi y s a ins and LDL-C le els. Simila ly, he e was
no associa ion de ec ed be ween he p esc ibing o high-in ensi y s a ins and age, smoking s a us, o
he p esence o hype ension o diabe es.
Figu e 1.
Numbe o wo ke s p esc ibed low–mode a e- o high-in ensi y s a ins, s a i ied by ca dio ascula
disease (CVD) isk and ecommended managemen . CVD isk: classi ica ion o ca dio ascula isk acco ding
o cu en Eu opean guidelines on CVD p e en ion, based on he combina ion o SCORE alue and he
p esence o o he CVD isk ac o s. Recommended managemen : ecommended pa ien managemen
acco ding o cu en Eu opean guidelines on CVD p e en ion, based on CVD isk, cu en LDL-choles e ol
(LDL-C) le els, and a ge LDL-C le els. NI: d ug ea men “no indica ed”, only li es yle ad ice
ecommended. <50%: d ug ea men indica ed o educe LDL-C by <50% o each a ge LDL-C
le els. ≥50%: d ug ea men indica ed o educe LDL-C by ≥50% o each a ge LDL-C le els.
Table 3.
Associa ion be ween sociodemog aphic, analy ical, clinical, and li es yle ac o s and he
p esc ibing o a high-in ensi y s a in in he s udy popula ion: logis ic eg ession analyses (N =569).
Unadjus ed Model Mul i a iable Model
OR (95%CI) OR (95%CI)
Age (yea s)
<50 0.90 (0.58–1.39) 1.03 (0.65–1.62)
50–55 1.00 1.00
55–60 0.88 (0.60–1.29) 0.90 (0.60–1.34)
>60 0.72 (0.25–2.09) 0.63 (0.21–1.90)
LDL-choles e ol (mg/dL)
Non-diabe ics
<155 1.00 1.00
≥155 1.04 (0.73–1.49) 1.03 (0.72–1.49)
Diabe ics
<155 1.00 1.00
≥155 4.50 (1.19–16.99) 4.96 (1.29–19.10)
Hype ension
No 1.00 1.00
Yes 1.25 (0.90–1.74) 1.26 (0.89–1.79)
Diabe es
No 1.00 1.00
Yes 1.42 (0.78–2.56) 1.45 (0.78–2.69)
J. Clin. Med. 2020,9, 3850 8 o 12
Table 3. Con .
Unadjus ed Model Mul i a iable Model
OR (95%CI) OR (95%CI)
Smoking
Ne e 1.00 1.00
Cu en 1.41 (0.88–2.25) 1.24 (0.74–2.06)
Fo me 1.20 (0.76–1.90) 1.13 (0.70–1.83)
METs h/wk
<20 1.00 1.00
20–40 0.90 (0.60–1.35) 0.93 (0.62–1.41)
>40 1.55 (1.03–2.32) 1.65 (1.08–2.50)
Ca o id a he oscle osis
No 1.00 1.00
Yes 1.33 (0.96–1.86) 1.21 (0.85–1.71)
Femo al a he oscle osis
No 1.00 1.00
Yes 1.53 (1.08–2.17) 1.45 (0.98–2.13)
Unadjus ed Model: logis ic eg ession model conside ing high-in ensi y s a in as ou come
and each indi idual a iable as exposu e. Adjus ed Model: logis ic eg ession model
adjus ed o age, smoking habi , LDL-choles e ol, diabe es, hype ension, METs h/wk,
and a he oscle osis. Abb e ia ions: OR, odds a io; 95%CI, 95% con idence in e al;
LDL-choles e ol, low-densi y lipop o ein choles e ol; METs h/wk, Me abolic equi alen s o
ask hou s pe week. S a is ically signi ican di e ences (p<0.05) a e shown in bold.
We conduc ed p edic i e models in o de o know which a iables p edic he p esc ibing o
high-in ensi y s a ins. The bes model ob ained included he a iables LDL-C, diabe es, hype ension,
smoking, and age (AUC, 0.620; 95%CI, 0.574–0.666). None heless, almos 40% o a iabili y emained
unexplained (Figu e 2).
J. Clin. Med. 2020, 9, x FOR PEER REVIEW 9 o 13
Figu e 2. P esc ibing o high-in ensi y s a ins: p edic i e capaci y o clinical and analy ical a iables.
Da a a e ep esen ed as he a ea unde he ecei e ope a ing cha ac e is ic (ROC) cu e and 95%
con idence in e als (e o ba s). Abb e ia ion: LDL-C, low-densi y lipop o ein choles e ol.
4. Discussion
The p esen s udy examines he sociodemog aphic, clinical, and li es yle ac o s ha de e mine
he ini ial p esc ibing o high-in ensi y s a ins in clinical p ac ice wi h he objec i e o achie ing he
LDL-C a ge le els ecommended in Eu opean guidelines on CVD p e en ion. Ou esul s indica e
ha while he choice o s a in ea men in ensi y is associa ed wi h ce ain CVD isk ac o s,
including LDL-C le els in diabe ics, in a la ge p opo ion o cases i appea s no o be based on he
clinical guidelines c i e ia.
Low–mode a e-in ensi y s a ins we e p esc ibed sligh ly mo e han high-in ensi y s a ins,
ega dless o basal CVD isk and he need o LDL-C educ ion o each he a ge le els
ecommended in cu en guidelines. Thus, se e al wo ke s who did no equi e a la ge dec ease in
LDL-C le els we e p esc ibed high-in ensi y s a ins, and some wo ke s who equi ed LDL-C
educ ions o o e 50% o each a ge le els began ea men wi h low–mode a e-in ensi y s a ins.
S ikingly, mo e han 10% o wo ke s included in ou s udy began ea men wi h s a ins (almos hal
wi h high-in ensi y s a ins) despi e he absence o any such indica ion in he Eu opean guidelines.
In ou s udy popula ion, he p obabili y o ecei ing high-in ensi y s a ins was highe in wo ke s
wi h diabe es and high LDL-C le els, in hose who pe o med mo e physical ac i i y, and in hose
wi h subclinical a he oscle osis in he emo al a e y. In ac , a g ea e p opo ion o wo ke s being
ea ed wi h high-in ensi y s a ins had ca o id and emo al a he oscle osis, al hough hese di e ences
we e no main ained a e adjus ing o o he con ounding a iables.
Macías Sain -Ge ons e al. [10] epo ed an associa ion be ween e y high LDL-C le els and he
p obabili y o p esc ibing high-in ensi y s a ins o p ima y CVD p e en ion. In ha s udy, which
included da a om almos 70,000 i s - ime use s be ween 2007 and 2011 in Spain, he au ho s also
de ec ed an associa ion be ween he p esc ibing o high-in ensi y s a ins and male sex, high BMI, and
smoking. Ou s udy he e o e suppo s a possible ole o LDL-C le els, among o he s, in he decision
o p esc ibe high-in ensi y s a ins. This makes sense since high-in ensi y s a ins educe LDL-C le els
o a g ea e ex en han he o he s. On he o he hand, diabe ic pa ien s ha e app oxima ely wice he
Figu e 2.
P esc ibing o high-in ensi y s a ins: p edic i e capaci y o clinical and analy ical a iables.
Da a a e ep esen ed as he a ea unde he ecei e ope a ing cha ac e is ic (ROC) cu e and 95%
con idence in e als (e o ba s). Abb e ia ion: LDL-C, low-densi y lipop o ein choles e ol.
J. Clin. Med. 2020,9, 3850 9 o 12
4. Discussion
The p esen s udy examines he sociodemog aphic, clinical, and li es yle ac o s ha de e mine he
ini ial p esc ibing o high-in ensi y s a ins in clinical p ac ice wi h he objec i e o achie ing he LDL-C
a ge le els ecommended in Eu opean guidelines on CVD p e en ion. Ou esul s indica e ha while
he choice o s a in ea men in ensi y is associa ed wi h ce ain CVD isk ac o s, including LDL-C le els
in diabe ics, in a la ge p opo ion o cases i appea s no o be based on he clinical guidelines c i e ia.
Low–mode a e-in ensi y s a ins we e p esc ibed sligh ly mo e han high-in ensi y s a ins,
ega dless o basal CVD isk and he need o LDL-C educ ion o each he a ge le els ecommended
in cu en guidelines. Thus, se e al wo ke s who did no equi e a la ge dec ease in LDL-C le els we e
p esc ibed high-in ensi y s a ins, and some wo ke s who equi ed LDL-C educ ions o o e 50% o
each a ge le els began ea men wi h low–mode a e-in ensi y s a ins. S ikingly, mo e han 10% o
wo ke s included in ou s udy began ea men wi h s a ins (almos hal wi h high-in ensi y s a ins)
despi e he absence o any such indica ion in he Eu opean guidelines.
In ou s udy popula ion, he p obabili y o ecei ing high-in ensi y s a ins was highe in wo ke s
wi h diabe es and high LDL-C le els, in hose who pe o med mo e physical ac i i y, and in hose wi h
subclinical a he oscle osis in he emo al a e y. In ac , a g ea e p opo ion o wo ke s being ea ed
wi h high-in ensi y s a ins had ca o id and emo al a he oscle osis, al hough hese di e ences we e no
main ained a e adjus ing o o he con ounding a iables.
Mac
í
as Sain -Ge ons e al. [
10
] epo ed an associa ion be ween e y high LDL-C le els and
he p obabili y o p esc ibing high-in ensi y s a ins o p ima y CVD p e en ion. In ha s udy,
which included da a om almos 70,000 i s - ime use s be ween 2007 and 2011 in Spain, he au ho s
also de ec ed an associa ion be ween he p esc ibing o high-in ensi y s a ins and male sex, high BMI,
and smoking. Ou s udy he e o e suppo s a possible ole o LDL-C le els, among o he s, in he
decision o p esc ibe high-in ensi y s a ins. This makes sense since high-in ensi y s a ins educe LDL-C
le els o a g ea e ex en han he o he s. On he o he hand, diabe ic pa ien s ha e app oxima ely
wice he isk o CVD han non-diabe ics [
24
]. In pa ien s wi h his combina ion o isk ac o s,
achie ing lowe LDL-C a ge le els is equi ed [
1
]. An a emp o compensa e hei excess o CVD isk
may be made by p esc ibing a high-in ensi y s a in. Thus, ou s udy esul s sugges ha p esc ibe s may
be guided by simple indica ions, such as ha he p esence o he combina ion o high
LDL-C +diabe es
,
which equi es a mo e in ense ea men .
In ou coho , high-in ensi y s a ins we e also mo e equen ly p esc ibed o wo ke s who
pe o med mo e physical ac i i y. Ho e al. [
11
], using da a om he Aus alian Diabe es, Obesi y and
Li es yle S udy, conduc ed a c oss-sec ional analysis o pa ien s aking s a ins and ound ha hose
who pe o med insu icien physical ac i i y we e mo e likely o be aking high-in ensi y s a ins.
Al hough ou esul s di e wi h hose o Ho e al., i should be no ed ha he la e s udy popula ion
consis ed o indi iduals subjec s who we e al eady aking s a ins, while ou s includes only wo ke s
who began his ea men du ing he s udy pe iod. The egula p ac ice o physical exe cise seems
o be ela ed o a g ea e conce n o heal h. In his sense, and gi en ha we measu ed he p ac ice
o physical exe cise be o e s a in p esc ip ion, we could hink ha people wi h a high CVD isk bu
in e es ed in hei heal h s a us could y o compensa e o his isk by ca ying ou mo e physical
exe cise. When his i s a emp ailed, hey would ecei e high-in ensi y s a in he apy. On he
con a y, when he le el o physical exe cise is measu ed in pa ien s who a e al eady aking s a ins,
he opposi e e ec could be obse ed, ha is, subjec s wi h a mo e in ensi e ea men would show a
elaxed a i ude owa ds physical exe cise as hey conside hemsel es p o ec ed by d ugs.
Finally, he ole o he known CVD isk ac o s in p edic ing he p esc ibing o high-in ensi y
s a ins was lowe han expec ed. This inding sugges s ha doc o s o en op o p esc ibe high-in ensi y
s a ins based on o he unknown c i e ia, pe haps in pa due o he lack o clea and di ec indica ions in
he Eu opean guidelines [
1
] on when o p esc ibe high-in ensi y s a ins, in con as o cu en Ame ican
College o Ca diology/Ame ican Hea Associa ion (ACC/AHA) guidelines [2]. Acco ding o cu en
Eu opean guidelines, in o de o choose he ype o s a in o be p esc ibed, he doc o mus know