Clinical Medicine 25 (2025) 100295
Con en s lis s a ailable a ScienceDi ec
Clinical Medicine
jou nal homepage: h ps://www.sciencedi ec .com/jou nal/clinical-medicine
Re iew
Respi a o y effec s o elec onic ciga e e use in indi iduals who ne e
smoked: A sys ema ic e iew
G azia Caci
a
, A ielle Selya
b , c
,Giusy Ri a Ma ia La Rosa
d
,Lucia Spicuzza
c , d , e
,
Jaymin B. Mo ja ia
,Giulio Ge aci
h
, Ricca do Polosa
c , d , g , h , ∗
a
UOC MCAU, Uni e si y Teaching Hospi al “Policlinico-S.Ma co ”, Uni e si y o Ca ania, I aly
b
Tobacco Ha m Reduc ion, Pinney Associa es, Inc., Pi sbu gh, PA, USA
c
Cen e o Excellence o he Accele a ion o HA m Reduc ion (CoEHAR), Uni e si y o Ca ania, Ca ania, I aly
d
Depa men o Clinical & Expe imen al Medicine, Uni e si y o Ca ania, Ca ania, I aly
e
Respi a o y Uni - Uni e si y Teaching Hospi al “Policlinico-S.Ma co ”, Uni e si y o Ca ania, Ca ania, I aly
Depa men o Respi a o y Medicine, Ha efield Hospi al, Guy’s & S Thomas’ NHS Founda ion T us , Ha efield, UK
g
Ins i u e o In e nal Medicine, Uni e si y Teaching Hospi al “Policlinico-S.Ma co ”, Uni e si y o Ca ania, Ca ania, I aly
h
Facul y o Medicine and Su ge y, “Ko e ”Uni e si y o Enna, Ko e, I aly
a i c l e i n o
Keywo ds:
e-ciga e es
lung unc ion
espi a o y symp oms
sys ema ic e iew
a b s a c
Cu en e idence on whe he elec onic ciga e es (ECs) pose espi a o y isks is unclea , due o con ounding by
ciga e e smoking; e idence among ne e -smoking indi iduals is needed. Following a na a i e e iew and c i -
ical app aisal, a sys ema ic e iew assessed possible espi a o y ou comes p ospec i ely associa ed wi h EC use
among indi iduals who ne e smoked. Bias isk was e alua ed using a Joanna B iggs Ins i u e ool. Ten eligible
s udies examined ou comes o sel - epo ed espi a o y diagnosis, symp oms and lung unc ion. Eigh examined
adul s and h ee examined you h (wi h o e lap). O e all, se en s udies showed no significan associa ion be ween
espi a o y ou comes and EC use among ne e -smoking indi iduals ( P > 0.05). E idence o coughing and wheez-
ing symp oms a ied by model specifica ion. O e all, EC use by ne e -smoking indi iduals is no associa ed wi h
isk o se e e espi a o y ou comes, bu may be associa ed wi h mild coughing/wheezing. Fu he esea ch is
needed using la ge samples, long- e m ollow-ups ( > 5 yea s), and in o ma ion on de ailed pa e ns o EC use.
Key p ac ice implica ions
Exclusi e use o elec onic ciga e es does no appea o be asso-
cia ed wi h se e e espi a o y isks, bu may pose a isk o mild
coughing and wheezing. E-ciga e es should be conside ed as a
ha m educ ion ool among adul s who smoke and a e unlikely o
qui .
In oduc ion
Elec onic ciga e es (ECs) a e inc easingly used o smoking cessa-
ion and ha m educ ion among adul s who smoke,
1–3 as hey mimic
he smoking expe ience wi hou he p oduc ion o ha m ul combus-
Abb e ia ions: AHR, adjus ed haza d a e; AOR, adjus ed odds a io; ARR, adjus ed isk a io; CI, confidence in e al; COPD, ch onic obs uc i e pulmona y
disease; EC, elec onic ciga e e; HRCT, high- esolu ion compu ed omog aphy; PATH, Popula ion Assessmen o Tobacco and Heal h; PICO, popula ion in e en ion
compa a o ou comes; PRISMA, P e e ed Repo ed I ems o Sys ema ic Re iews and Me a-Analyses; RCT, andomised con olled ial; VERITAS S udy, he Vaping
Effec s-Real-Wo ld In e na ional Su eillance S udy.
∗ Co esponding au ho .
E-mail add ess: [email p o ec ed] (R. Polosa) .
ion o smoke.
4 , 5 While ECs p oduce significan ly lowe exposu es o
ha m ul subs ances han obacco ciga e es
6–8
and a e associa ed wi h
lowe nico ine dependence le els,
9 , 10 conce ns emain abou depen-
dence
11 and heal h effec s esul ing om long- e m EC use,
12 espe-
cially by you h and obacco-naï e indi iduals. EC oxici y likely a ies
by p oduc -specific cha ac e is ics, including fla ou ings, hough again,
being non-combus ible p oduc s, ECs a e ca ego ically less oxic.
12 , 13
While da a a e no ye a ailable o heal h ou comes equi ing long-
e m cumula i e exposu es (eg ∼30 yea s o lung cance ), espi a o y
ou comes can plausibly de elop o e sho - o medium- e m du a ions
(conside ed he e as ∼1–5 yea s). P e ious sys ema ic and/o scoping
e iews ha e epo ed ha EC use is associa ed wi h lowe oxic expo-
su es han ciga e es
11 , 13–15
bu ha he e a e some isks o espi a o y
i i a ion,
15
mild ad e se e en s
11
and acu e espi a o y changes.
13 , 14
h ps://doi.o g/10.1016/j.clinme.2025.100295
Recei ed 30 Oc obe 2024; Recei ed in e ised o m 15 Janua y 2025; Accep ed 14 Feb ua y 2025
1470-2118/© 2025 The Au ho (s). Published by Else ie L d on behal o Royal College o Physicians. This is an open access a icle unde he CC BY license
( h p://c ea i ecommons.o g/licenses/by/4.0/ )
G. Caci, A. Selya, G.R.M. La Rosa e al. Clinical Medicine 25 (2025) 100295
Howe e , none o hese p e ious e iews obus ly accoun ed o he con-
ounding effec o p io ciga e e smoking o ocused specifically on in-
di iduals wi hou an es ablished smoking his o y. Fu he , he acu e es-
pi a o y effec s ha e unknown clinical significance, making he unique,
clinical espi a o y effec s o ECs unclea om exis ing e iews.
Focusing on indi iduals wi hou es ablished smoking habi s (’ne e -
smoking’), he e o e, is pa icula ly in o ma i e o examining possible
espi a o y isks unique o EC use, as i a oids con ounding by ciga e e
smoking his o y. Howe e , EC use by ne e -smoking indi iduals is ela-
i ely a e.
16–19
Al hough he p opo ion o EC use s who a e smoking-
naï e appea s o be inc easing as popula ion-le el nico ine consump ion
shi s om smoking o aping, especially among younge age g oups,
20
his g oup ne e heless con ibu es e y li le o he e idence base a
p esen . Acco dingly, ou g oup p e iously pe o med a na a i e e-
iew and c i ical app aisal ocusing on e idence among ne e -smoking
adolescen s and adul s,
21 concluding ha he e is some e idence o
coughing o wheezing symp oms bu ha EC use is unlikely o pose
significan o clinically meaning ul espi a o y ha ms o e he medium
e m. Howe e , hese s udies had impo an limi a ions, including an
o e - ep esen a ion o US da a (especially om he Popula ion Assess-
men o Tobacco and Heal h (PATH)), inadequa e con ols o con ound-
ing (eg o he combus ible obacco use), and limi ed ollow-up du a ions.
He e, we ex end his p io na a i e e iew
21
h ough a o mal sys-
ema ic e iew aimed a syn hesising exis ing e idence on possible espi-
a o y ou comes p ospec i ely associa ed wi h EC use in ne e -smoking
indi iduals. Due o he small numbe o quali ying s udies, we consid-
e ed all a ailable espi a o y ou comes (sel - epo ed diagnoses, espi-
a o y symp oms and lung unc ion es s).
Ma e ial and me hods
Seach s a egy and selec ion c i e ia
This sys ema ic e iew was epo ed ollowing he P e e ed Re-
po ed I ems o Sys ema ic Re iews and Me a-analyses (PRISMA)
guidelines
22 and he p o ocol was egis e ed a p io i in PROSPERO
(CRD42024554721).
Ou sea ch used he ollowing PICO (Popula ion, In e en ion, Com-
pa a o , Ou come) amewo k: P: Adul s ( ≥ 18 yea s) and you h (12–
17 yea s) wi h no es ablished use o combus ible ciga e es (ie ne e -
smoking
1
); I: Cu en EC use; C: Non-cu en EC use (wi h o me - s.
ne e -EC use combined o sepa a e); O: Any sel - epo ed o clinically-
alida ed modifica ion in espi a o y unc ion, including sel - epo ed
diagnosis (eg p e alence o incidence/onse ) o espi a o y disease (eg
as hma); espi a o y symp oms (p e alence, equency o exace ba ion),
o lung unc ion es s.
A comp ehensi e sea ch in PubMed and Scopus was conduc ed in
May 2024 using he keywo ds ’ne e smok∗
’, ’näi e’, ’e-cig∗
’, ’ espi ∗
’
and ’as hma’ (see Supplemen a y Table S1). The e e ence lis s o in-
cluded a icles and e iew pape s we e also sc eened. Expe s in obacco
ha m educ ion and/o espi a o y unc ion we e consul ed o confi m
ha all pe inen s udies we e included.
We included English-language s udies wi h p ospec i e designs only
(clinical obse a ional s udies, andomized clinical ials (RCTs), and
popula ion su eys) o ensu e he co ec empo al sequence.
This e iew excluded s udies ha did no dis inguish o me -smoking
om ne e -smoking, as linge ing effec s om p io smoking could con-
ound esul s. We also excluded labo a o y s udies ha measu ed acu e
exposu es o EC ae osol, as hese ou comes p ima ily documen mild,
ansien espi a o y symp oms (eg h oa i i a ion, cough, wheezing,
o modes changes in espi a o y physiology and ai way inflamma-
1 “Ne e -smoking ”as used he e includes bo h ne e - es ablished smoking (i.e.,
< 10 + ciga e es/li e ime, as ypically assessed in adul s) and ue ne e -smoking
(i.e., no e en a puff; ypically assessed in you h).
ion
23–25
) wi h unclea clinical ele ance. Simila ly, cellula and ani-
mal labo a o y s udies we e excluded due o limi ed ele ance o hu-
man clinical heal h
26
and use o un ealis ic ope a ing condi ions o EC
de ices which o e s a e nega i e impac s and lack app op ia e expe i-
men al con ols.
27 , 28
S udies o hea ed obacco p oduc s we e excluded,
as we e e iews, s udy p o ocols, case epo s and con e ence abs ac s.
Two e iewe s (GRMLR and GC) independen ly sc eened he i les
and abs ac s o all iden ified eco ds o de e mine which s udies e-
qui ed a ull- ex e iew; hese we e e ie ed o u he e alua ion.
The wo e iewe s and AS hen independen ly assessed each ull ex
o de e mine final eligibili y. Any disag eemen s we e esol ed h ough
discussion and consensus wi h an addi ional e iewe (RP).
Da a analysis
Da a ex ac ion and abula ion o each included s udy was ca ied
ou independen ly by wo au ho s (GRMLR and AS), and he ollow-
ing in o ma ion was abula ed: fi s au ho , yea , coun y, s udy design,
popula ion, size (N) o cu en EC use g oup (ie among ne e -smoking
indi iduals), size (N) o non-cu en EC use g oup (which may o may
no dis inguish o me - and ne e -EC use), smoking beha iou e ifica-
ion, ou come, ollow-up du a ion, main esul s, conclusions and und-
ing.
A me a-analysis was no easible due o he e ogenei y in ou comes
and he ype o s a is ical es ima e ac oss s udies (eg diagnoses s. symp-
oms; p e alence s. incidence) and he e y small numbe o s udies
( < 5) in each compa able g oup.
Risk o bias assessmen was pe o med independen ly by wo au-
ho s (GRMLR and GC) using he Joanna B iggs Ins i u e’s c i ical ap-
p aisal ools app op ia e o he s udy design ( h ps://jbi.global/c i ical-
app aisal- ools ). Any disag eemen was esol ed h ough consensus dis-
cussions o , i necessa y, by consul ing a hi d au ho (RP).
Resul s
Ten s udies me inclusion c i e ia ( Fig. 1 ). The lis o excluded s udies
a e ull- ex e iew is p o ided in Supplemen a y Table S2.
Table 1 p esen s he main cha ac e is ics o he included s udies.
Eigh s udies analysed adul s (o which h ee ocused only on younge
adul s, < 24 o < 30), and h ee s udies analysed you h (one s udy ex-
amined bo h you h and adul s). All s udies defined EC use as cu en
use (ei he ’some days’ o ’e e y day’ s. ’no a all’ o any pas -30-day
use). In six s udies, he compa ison g oup was non-cu en EC use (ie
combining o me - and ne e -EC use), while he emaining ou s udies
dis inguished o me - and ne e -EC use. Mos s udies (n = 8) analysed
da a om he US Popula ion Assessmen o Tobacco and Heal h (PATH)
S udy, one s udy analysed linked da a be ween a Canadian popula ion
su ey and adminis a i e heal h eco ds, and one s udy ec ui ed a
small coho in I aly.
Wi h espec o espi a o y ou comes, ou s udies examined sel -
epo ed espi a o y diagnosis, wi h wo examining p e alence (one
o any espi a o y disease, one o as hma) and wo examining inci-
dence/onse o as hma (wi h one examining only age o onse ). Six s ud-
ies examined sel - epo ed espi a o y symp oms (mos o en wheezing;
h ee used a h eshold o unc ionally impo an symp oms). One s udy
examined lung unc ion ou comes based on spi ome y es s and high-
esolu ion compu ing omog aphy (HRCT).
As a me a-analysis was no easible (see Me hods), we quali a i ely
syn hesise he findings, o ganising by ype o ou come. Fi s , o he
ou pape s examining sel - epo ed diagnosis, h ee ocused on as hma
specifically. To e al
37 ound no associa ion be ween cu en ( s. non-
cu en ) EC use and p e alence o sel - epo ed as hma (adjus ed odds
a io (AOR) = 1.21 [95% confidence in e al (CI) [0.95–1.54]) among a
sample o Canadian younge adul s (ages 15–30) o pas -yea as hma a -
acks among hose wi h as hma.
38
The e was an in e ac ion wi h sex on
pas -yea as hma a acks , such ha emale EC use s ( s. male non-use s)
2
G. Caci, A. Selya, G.R.M. La Rosa e al. Clinical Medicine 25 (2025) 100295
Fig. 1. Sea ch s a egy.
had highe odds o ha ing an as hma a ack (AOR = 2.30[1.29–4.12]);
howe e , he e was a simila diffe ence be ween emale s. male EC
use s (AOR = 2.29[1.57–3.35]), so i is unclea whe he his esul is a -
ibu able o EC use o a sex diffe ence. We also no e ha To e al adjus ed
o smoking s a us (using ne e -smoking as he e e ence g oup) a he
han excluding o me ly- and cu en ly-smoking adul s en i ely; hus,
esul s may no gene alise o ne e -smoking adul s specifically (see ou
na a i e e iew
21
).
Pa el e al
31
and Pe ez e al
32
bo h analysed ou comes o as hma inci-
dence (ie new onse since baseline) using PATH da a. Pa el e al
31
ound
ha EC use among ne e -smoking you h (ages 12–17) was no signifi-
can ly associa ed wi h onse o as hma 1 yea la e (adjus ed haza d a e
(AHR) = 1.25[0.77–2.04]). Simila ly, Pe ez e al
32
did no find baseline
EC use o be associa ed wi h age o as hma onse among you h naï e
o smoking and as hma baseline (AHR = 1.55[0.60–3.96]), hough he
associa ion was significan among adul s (age 18 + ) (AHR = 3.66[1.23–
10.85]). Howe e , he la e became non-significan in a supplemen a y
analysis ha excluded pa icipan s wi h any baseline combus ible o-
bacco p oduc use (AHR = 1.45[0.14–15.04]; see Discussion).
Finally, Kenkel e al
29 pe o med a eplica ion and ex ension o
Bha a & Glan z’s PATH analysis
30 o EC use and sel - epo ed diag-
noses o any espi a o y disease (COPD, ch onic b onchi is, emphysema
o as hma). Unlike he o iginal a icle, Kenkel e al
29
diffe en ia ed be-
ween o me - and ne e -smoking, and ound no significan associa ion
be ween exclusi e EC use and p e alence o espi a o y disease o e a
3-yea pe iod ( hough see Discussion o imp ecision due o low num-
be s).
Toge he , hese s udies o e all ail o suppo ha exclusi e EC use
is associa ed wi h espi a o y diagnoses among indi iduals who ne e
smoked.
Six s udies examined espi a o y symp om ou comes. Polosa e al
33
assessed espi a o y symp oms (cough, wheezing, sho ness o b ea h,
and igh ches ) p ospec i ely o e an a e age o 3.5 yea s among a
small (N = 21) coho o adul EC use s and non-EC-using con ols; he e
was no e idence o espi a o y symp oms om EC use alone (only one
symp om, cough, was epo ed by wo EC use s and h ee con ols; s a is-
ical es no pe o med due o small numbe s). The emaining fi e s ud-
ies all analysed PATH da a. In ne e -smoking adul s, Sanchez-Rome o
e al
34 ound ha EC use a baseline was no significan ly associa ed
wi h highe odds o wheezing o e 5 yea s (AOR = 1.20[0.83–1.72]), and
Ka ey e al.
40
ound no associa ion wi h unc ionally impo an espi a-
o y symp oms, using a h eshold on a six-i em index on wheezing and
nigh - ime d y cough (AOR = 0.82[0.27–2.56]).
Sa gen e al
35
and Xie e al
39
bo h examined sel - epo ed espi a-
o y symp oms in young adul s (ages 18–24) in PATH. Sa gen e al
35
examined unc ionally impo an espi a o y symp oms, defined using
wo diffe en cu -offs (o 2 + and 3 + on a 0–9 scale) on a se en-i em
espi a o y symp om index, as well as wo sening and imp o emen o
hese symp oms o e ime. Baseline cu en EC use ( s. ne e -use) was
no associa ed wi h diffe en likelihood o unc ionally impo an es-
pi a o y symp oms bu was, in some bu no all models, significan ly
associa ed wi h wo sening o symp oms (ie significan when using he
2 + cu -off (adjus ed isk a io (ARR) = 1.63[1.02–2.59]), bu no 3 +
(ARR = 1.58[0.84–2.96]). Findings on symp om imp o emen also a ied
in di ec ion depending on he cu -off alue (see Table 1 ). Like To e
al ,
37
Sa gen e al
35
es ima es may be biased, as analyses adjus ed o
smoking s a us (using ne e -smoking as he e e ence g oup) a he han
omi ing o me and cu en smoking g oups en i ely (see ou na a i e
e iew
21
). In con as , Xie e al.
39 who use he same da a sou ce and
age ange abo e, epo ha cu en ( s. ne e ) EC use a baseline was
associa ed wi h highe odds o subsequen wheezing symp oms (’in he
ches ’: AOR = 2.23[1.28–3.91]; ’du ing exe cise’: AOR = 2.41[1.39–4.16])
and any espi a o y symp om (AOR = 1.86[1.35–2.58]). Howe e , hese
associa ions migh eflec con ounding by o he obacco use (see Discus-
sion).
S e ens e al
36
examined you h (ages 12–17) in PATH and ou comes
o sel - epo ed espi a o y symp oms, using a se en-i em index wi h a
h eshold o 2 + on a 0–9 scale, simila o Ka ey e al
40
and Sa gen e al
35
abo e. Pas 30-day EC use ( s. non-use) a baseline was no significan ly
associa ed impo an espi a o y symp oms 1 yea la e among ne e -
smoking you h (AOR = 0.86[0.32–2.32]).
O e all, hese six s udies show somewha mixed e idence: h ee
s udies (Ka ey e al
40
; Sanchez-Rome o e al
34 and S e ens e al
36
) e-
po ed a non-significan associa ion wi h EC use; one (Polosa e al
33
) e-
po ed some coughing symp oms in bo h EC use s and non-use s, wi h
unknown s a is ical significance; and wo s udies (Pe ez e al
32
and Sa -
gen e al
35
) epo ed enuous associa ions whose s a is ical significance
depended on model specifica ions.
Finally, Polosa e al
33
was he only s udy o examine objec i e ou -
comes using lung unc ion es s, bioma ke s o ai way inflamma ion (ex-
3
G. Caci, A. Selya, G.R.M. La Rosa e al. Clinical Medicine 25 (2025) 100295
Table 1
Gene al cha ac e is ics o he included s udies.
Fi s au ho , yea ,
coun y s udy design Popula ion
Cu en e-cig use
(ne e smoke s) (N)
Non-cu en e-cig
use (Fo me o /and
ne e use among
ne e smoke s) (N)
Smoking
beha iou
e ifica ion Ou come Follow-up Main esul s Conclusions Funding
Ka ey
e al
23
USA
Na ionally
ep esen a i e,
longi udinal coho
su ey (PATH)
Adul s
( ≥ 18 yea s)
N = 65
Any P30D
e-ciga e e use
Ne e e-cig use
N = 3,323
Fo me e-cig use
N = 250
Sel - epo ed Respi a o y
symp oms index
(based on six
wheezing i ems and
a nigh - ime d y
cough i em wi h
highe alues
indica ing mo e
espi a o y
symp oms)
2 yea s
Wa es 4
(2016–2018)
and 5
(2018–2019)
Re e ence: Ne e use
AOR = 1
Cu en e-cig use:
AOR = 0.82
95% CI [0.27–2.56]
P = 0.736
No significan
associa ion be ween
e-cig use and impo an
espi a o y symp oms
among ne e smoke s
was de ec ed.
NR
Kenkel e al
29
USA
Na ionally
ep esen a i e,
longi udinal coho
su ey (PATH) based
on ex ension analysis
o Bha a and
Glan z’s
30
esul s
Adul s (aged
≥ 18 yea s)
N = 12
E e used an
e-ciga e e ’ ai ly
egula ly’ and
cu en ly used
e-ciga e es e e y
day o some days
Ne e use
N = 2,705
Fo me use
N = 51
Sel - epo ed Lung o espi a o y
disease (sel - epo s
o whe he hey had
e e been old hey
ha e COPD, ch onic
b onchi is,
emphysema o
as hma)
1 yea
Wa es 1
(2013–2014)
and 3
(2015–2016)
Re e ence: Ne e use
Coeff. LPM =− 0.03
95% CI [− 0.04, − 0.01]
The s a is ical so wa e d opped
da a o he cu en e-cig
use s/ne e smoke s om he
model because he indica o o
his ca ego y pe ec ly p edic ed
he ou come –all 12 pa icipan s
did no epo inciden
espi a o y diseases.
a
Among ne e smoke s,
he e was no e idence
ha cu en e-cig use
was associa ed wi h
espi a o y disease.
Co nell
Uni e si y
Pa el e al
31
USA
Na ionally
ep esen a i e,
longi udinal coho
su ey (PATH)
You hs (aged
12–17 yea s)
N = 142
F = 54
M = 88
Any P30D
e-ciga e e use
Non-cu en use
(N = 8,590)
F = 4,346
M = 4,244
Sel - epo ed As hma incidence
(sel - epo ed)
5 yea s
S udy Wa es
1–5
(2013–2019)
Re e ence: Non-cu en use
HR = 1
Exclusi e ENDS use
Unadjus ed
HR = 1.19
95% CI [0.73–1.96]
P = 0.477
Adjus ed
a
HR = 1.25
95% CI [0.77–2.04]
P = 0.359
Sho - e m exclusi e
ENDS use was no
s a is ically associa ed
wi h highe isk o
inciden diagnosed
as hma o e 5 yea s.
Na ional Cance
Ins i u e (NCI)
o he Na ional
Ins i u es o
Heal h (NIH)
and he Food
and D ug
Adminis a ion
(FDA) Cen e o
Tobacco
P oduc s.
( con inued on nex page )
4
G. Caci, A. Selya, G.R.M. La Rosa e al. Clinical Medicine 25 (2025) 100295
Table 1 ( con inued )
Fi s au ho , yea ,
coun y s udy design
Popula ion Cu en e-cig use
(ne e smoke s) (N)
Non-cu en e-cig
use (Fo me o /and
ne e use among
ne e smoke s) (N)
Smoking
beha iou
e ifica ion
Ou come Follow-up Main esul s Conclusions Funding
Pe ez e al
32
USA
Na ionally
ep esen a i e,
longi udinal coho
su ey (PATH)
Adul s (aged
≥ 18 yea s) and
you hs (aged
12–17 yea s)
Main analysis (based
on fi s wa e)
Adul s
N = 160
You hs
N = 96
Sensi i i y analysis:
Adul s
N = 62
You hs
N = 59
Any P30D e-ciga e e
No P30D ENDS use
(ne e smoke s a
wa e 1)
Main analysis (based
on fi s wa e)
Adul s
N = 7,606
You hs
N = 16,927
Sensi i i y analysis:
Ne e use
Adul s
N = 5,355
You hs
N = 15,394
Sel - epo ed As hma onse Median (SE)
Adul s: 4.94
(0.06) yea s
You hs: 4.19
(0.04) yea s
Wa es 1
(2013–2014) o
6 (2020–2021)
Main analysis: among
pa icipan s epo ed ne e
smoking a fi s wa e:
Re e ence: Non-cu en ENDS use
Adul s
C ude associa ion
HR = 9.57
95% CI [3.76–24.33]
Model 3
HR = 3.66
95% CI [1.23–10.85]
You hs
C ude associa ion
HR = 1.18
95% CI [0.46–2.70]
Model 3
HR = 1.55
95% CI [0.60–3.96]
Sensi i i y analysis:
Re e ence: Non-cu en ENDS use
and ne e -use o combus ible
obacco a fi s wa e
HR = 1
Cu en ENDS use a he fi s
wa e
Adul s
C ude associa ion
HR = 2.86
95% CI [0.36–22.82]
Model 1
AHR = 1.42
95% CI [0.14–14.59]
Model 2
AHR = 1.45
95% CI [0.14–15.04]
You hs
C ude associa ion
HR = 0.66
95% CI [0.10–4.24]
Model 1
AHR = 0.70
95% CI [0.10–4.83]
Model 2
AHR = 0.68
95% CI [0.10–4.65]
Adul s, bu no you hs,
who epo ed ne e
ciga e es and P30D
ENDS use a he fi s
wa e showed highe isk
o as hma incidence a
ea lie ages in
compa ison wi h hose
who epo ed no P30D
ENDS use.
The e was no associa ion
be ween he P30D ENDS
use wi h he age o
as hma onse among
adul s and you hs
compa ing use s and
ne e use (among ne e
smoke s a he fi s
wa e) in he sensi i i y
analysis.
Na ional Hea ,
Lung, and Blood
Ins i u e and he
US FDA Cen e
o Tobacco
P oduc s.
( con inued on nex page )
5
G. Caci, A. Selya, G.R.M. La Rosa e al. Clinical Medicine 25 (2025) 100295
Table 1 ( con inued )
Fi s au ho , yea ,
coun y s udy design
Popula ion Cu en e-cig use
(ne e smoke s) (N)
Non-cu en e-cig
use (Fo me o /and
ne e use among
ne e smoke s) (N)
Smoking
beha iou
e ifica ion
Ou come Follow-up Main esul s Conclusions Funding
Polosa e al
33
I aly
P ospec i e coho
s udy
Adul s
( ≥ 18 yea s)
N = 9
F = 3
M = 6
26.6 ± 6.0 yea s
Daily e-ciga e e
use s o ≥ 3 mon hs
Ne e e-cig use
N = 12
F = 4
M = 8
27.8 ± 5.2 yea s
Sel - epo ed Lung unc ion,
espi a o y
symp oms, eNO, eCO
and HRCT o he
lungs
3.5 yea s FEV1 (l, mean ± SD)
Baseline; F/up 3
EC use s: 3.82 ± 0.78; 3.87 ± 0.76
Con ol: 4.08 ± 0.30; 4.11 ± 0.30
P > 0.05
FVC (l, mean ± SD)
Baseline; F/up 3
EC use s: 4.93 ± 0.95; 4.87 ± 0.83
Con ol: 5.03 ± 0.48; 5.02 ± 0.42
P > 0.05
FEV1/FVC (%, mean ± SD)
Baseline; F/up 3
EC use s: 78.49 ± 3.46;
79.08 ± 2.83
Con ol: 81.45 ± 5.03; 82.06 ± 4.25
P > 0.05
FEF25–75% (l/min, mean ± SD)
Baseline; F/up 3
EC use s: 3.29 ± 0.70; 3.33 ± 0.64
Con ol: 3.43 ± 0.64; 3.56 ± 0.58
P > 0.05
eCO (ppm, median and IQ ange)
Baseline; F/up 3
EC use s: 5.0 [3.5–7.3]; 4.0
[2.8–6.3]
Con ol: 4.0 [3.5–7.5]; 5.0
[5.5–6.0]
P > 0.05
FeNO (ppb, median and IQ ange)
Baseline; F/up 3
EC use s: 21.1 [16.2–24.5]; 20.0
[18.2–22.7]
Con ol: 18.6 [17.6–25.7]; 20.0
[16.2–23.4]
P > 0.05
None o he pa icipan s e e ed
any wheezing, sho ness o
b ea h, o ches igh ness.
Cough = 2 EC use and 3 con ols.
No pa hological findings we e
iden ified on HRCT o he lungs.
This small s udy showed
no de ec able
modifica ions in lung
heal h in ne e smoke s
who ha e been egula ly
aping o a leas 4
yea s.
Uni e si y g an
( con inued on nex page )
6
G. Caci, A. Selya, G.R.M. La Rosa e al. Clinical Medicine 25 (2025) 100295
Table 1 ( con inued )
Fi s au ho , yea ,
coun y s udy design
Popula ion Cu en e-cig use
(ne e smoke s) (N)
Non-cu en e-cig
use (Fo me o /and
ne e use among
ne e smoke s) (N)
Smoking
beha iou
e ifica ion
Ou come Follow-up Main esul s Conclusions Funding
Sanchez-Rome o e
al
34
USA
Na ionally
ep esen a i e,
longi udinal coho
su ey (PATH)
Adul s
( ≥ 18 yea s)
% (SE)
Wa e 1
(2013–2014)
0.3 (0.03)
Wa e 2
(2014–2015)
0.3 (0.03)
Wa e 3
(2015–2016)
0.3 (0.04)
Wa e 4
(2016–2018)
0.3 (0.03)
Wa e 5
(2018–2019)
NA
N = 51
b
(ac oss wa es
1–4)
Cu en ly used
e-ciga e es on e e y
day o some days
Noncu en ENDS
use s
% (SE)
Wa e 1
(2013–2014)
61.8 (0.62)
Wa e 2
(2014–2015)
58.6 (0.64)
Wa e 3
(2015–2016)
57.1 (0.66)
Wa e 4
(2016–2018)
56.3 (0.66)
Wa e 5
(2018–2019)
NA
N =∼9500–10,500
b
(ac oss wa es 1–4)
Sel - epo ed Wheezing symp oms
(sel - epo ed)
5 yea s
Wa es 1
(Sep embe
2013 –
Decembe 2014)
o 5 (Decembe
2018 –
No embe 2019)
Re e ence: Non-cu en use
AOR = 1
Cu en ENDS use:
AOR = 1.20
95% CI [0.83, 1.72]
P = 0.0.32
Supplemen a y analysis using
3-le el ENDS use (ne e , o me ,
cu en ):
Re e ence: Ne e use
AOR = 1
Cu en ENDS use:
AOR = 1.28
95% CI [0.89, 1.85]
P = 0.19
Exclusi e ENDS use was
no significan ly
associa ed wi h an
inc ease in odds o
sel - epo ed wheezing
compa ed wi h ne e use
and non-cu en ENDS
use among ne e
smoke s.
NCI o he NIH
and FDA Cen e
o Tobacco
P oduc s
Sa gen e al
35
USA
Na ionally
ep esen a i e,
longi udinal coho
su ey (PATH)
Young adul s
(18–24 yea s)
N = 327 among
o me - and
ne e -smoke s (no
sepa a ely epo ed)
Any P30D
e-ciga e e use
N = 5,888
Non-cu en
e-ciga e e use
Sel - epo ed Respi a o y
symp om index
based on se en
wheezing/cough
ques ions om
ISAAC wi h highe
alues indica ing
mo e espi a o y
symp oms. Cu off
alues ≥ 2 and ≥ 3.
Wa es 1
(2013–2014) o
2 (2014–2015)
and 3
(2015–2016)
Re e ence: Ne e use
Unadjus ed
Exclusi e e-cig use
RR = 1.53
95% CI [0.98–2.40]
Wo sening symp oms
(asymp oma ic Wa e 2-
symp oma ic Wa e 3), adjus ing
o smoking his o y wi h
ne e -smoking as he e e ence
g oup:
Cu off≥ 2
RR = 1.63 95% CI [1.02, 2.59]
Cu off≥ 3
RR = 1.58 95% CI [0.84, 2.96]
Imp o emen symp oms
(symp oma ic Wa e 2-
asymp oma ic Wa e 3), adjus ing
o smoking his o y wi h
ne e -smoking as he e e ence
g oup:
Cu off≥ 2
RR = 0.57 95% CI [0.40, 0.82]
Cu off≥ 3
RR = 1.64 95% CI [1.04, 2.58]
Compa ed o ne e use s,
exclusi e use s o e-cigs
exhibi ed a simila isk
o unc ionally impo an
espi a o y symp oms,
independen ly om he
cu off conside ed.
In exclusi e e-cig use,
wo sening o symp oms
depended om he
symp oms se e i y and
cu off conside ed.
Fede al unds
om he
Na ional
Ins i u e on D ug
Abuse, NIH, and
he Cen e o
Tobacco
P oduc s, FDA,
Depa men o
Heal h and
Human Se ices
( con inued on nex page )
7
G. Caci, A. Selya, G.R.M. La Rosa e al. Clinical Medicine 25 (2025) 100295
Table 1 ( con inued )
Fi s au ho , yea ,
coun y s udy design
Popula ion Cu en e-cig use
(ne e smoke s) (N)
Non-cu en e-cig
use (Fo me o /and
ne e use among
ne e smoke s) (N)
Smoking
beha iou
e ifica ion
Ou come Follow-up Main esul s Conclusions Funding
S e ens e al
36
USA
Na ionally
ep esen a i e,
longi udinal coho
su ey (PATH)
You hs
(aged 12–17
yea s)
N = 54 (1.7%)
Any P30D
e-ciga e e use
Ne e e-cig use
N = 2,998 (90.8%)
Fo me e-cig use
N = 266
(7.5%)
Sel - epo ed Respi a o y
symp om index
(based on esponses
o se en wheezing
i ems wi h an index
o ≥ 2 indica ing
unc ionally
impo an
espi a o y
symp oms)
1 yea
Wa es 3
(Oc obe 2015 –
Oc obe 2016)
and 4
(Decembe 2016
–Janua y 2018)
Re e ence: Ne e use
AOR = 1
Fo me e-cig use:
AOR = 1. 20
95% CI [0.78, 1.85]
P = 0.411
Cu en e-cig use:
AOR = 0.86
95% CI [0.32, 2.32]
P = 0.767
E-ciga e e use
(including o me and
cu en ) was no
significan ly associa ed
wi h highe odds o a
espi a o y symp om a
1-yea ollow-up among
ne e combus ible
obacco use s.
NCI a he NIH
To e al
37
Canada
Na ionally
ep esen a i e,
longi udinal coho
su ey
CCHS and heal h
adminis a i e
da abases (DAD and
NACR)
Adul s
(aged 15–30
yea s)
N = 75
b
Any P30D
e-ciga e e use
N = 365
a , b
Non-cu en
e-ciga e e use
Sel - epo ed As hma p e alence,
as hma a acks
(sel - epo ed)
CCHS (cycles
2015–16 and
2017–18) and
heal h
adminis a i e
da a (Janua y
2015 –Ma ch
2018)
Re e ence: Non-cu en use
As hma p e alence: Cu en
ENDS use, adjus ing o smoking
his o y wi h ne e -smoking as
he e e ence g oup:
AOR = 1.21
95% CI [0.95, 1.54]
P = 0.1170
While he e was no
analysis specific o
ne e -smoke s, he
adjus ed model using
ne e -smoke s as he
e e ence g oup ound no
associa ion be ween
cu en ENDS use and
as hma p e alence o
as hma a acks.
38
The Canadian
Ins i u es o
Heal h Resea ch
Ca alys G an :
Heal h Effec s o
Vaping.
Xie e al
39
USA
Na ionally
ep esen a i e,
longi udinal coho
su ey (PATH)
Young adul s
(18–24 yea s)
N = 312
E e used an
e-ciga e e ’ ai ly
egula ly’ and
cu en ly used
e-ciga e es e e y
day o some days
N = 8,388
Ne e used
e-ciga e es
N = 1,140
Fo me ly used
e-ciga e es
Sel - epo ed Respi a o y
symp oms (ie
wheezing in he
ches , and du ing o
a e exe cise, d y
cough a nigh )
1 yea
Wa es 1
(2014–2015) o
5 (2018–2019)
wi h exposu e
wa e (wa es
2–4) and
ou come wa e
(wa es 3–5)
Re e ence: Ne e use
OR = 1
Exclusi e ENDS use
Any espi a o y symp om
Fully adjus ed
OR = 1.86
95% CI [1.35–2.58]
Wheezing in he ches
Fully adjus ed
OR = 2.23
95% CI [1.28–3.91]
Wheezing du ing exe cise
Fully adjus ed
OR = 2.41
95% CI [1.39–4.16]
D y cough a nigh
Fully adjus ed
OR = 1.41
95% CI [0.97–2.04]
Among ne e smoke s,
cu en e-cig use was
associa ed wi h 86%
highe odds o epo ing
any espi a o y symp om
han ne e use. The
associa ion was
pa icula ly s ong o
wheezing in he ches
and du ing exe cise.
Ame ican Lung
Associa ion
Public Policy
Resea ch Awa d,
NHLBI g an and
Ame ican Hea
Associa ion
Tobacco Cen e
o Regula o y
Science g an s
AOR, adjus ed odds; CCHS, Canadian Communi y Heal h Su ey; CI, confidence in e al; DAD, Discha ge Abs ac Da abase; ENDS, elec onic nico ine; eNO, exhaled b ea h ni ic oxide, eCO, exhaled ca bon
monoxide, F, emale; HR, Haza d Ra io; HRCT, high- esolu ion compu ed omog aphy; ISAAC, he In e na ional S udy o Alle gies and As hma in Childhood; LPM, linea p obabili y model; NACRS, Na ional
Ambula o y Ca e Repo ing Sys em; NCI, Na ional Cance Ins i u e; NR, No Repo ed; M, male; PATH, Popula ion Assessmen o Tobacco and Heal h; P30D, pas 30-day.
a Adjus ing o baseline age, sex, ace/e hnici y, pa en al educa ional a ainmen , u banici y, second-hand smoke exposu e, household combus ible obacco use, and BMI- o -age.
b N alues we e es ima ed om o he numbe s in he pape , as hey we e no di ec ly epo ed.
8
G. Caci, A. Selya, G.R.M. La Rosa e al. Clinical Medicine 25 (2025) 100295
haled b ea h ni ic oxide [eNO] and ca bon monoxide [eCO]) and HRCT.
In his p ospec i e obse a ional s udy o 31 ne e -smoking pa icipan s
(16 daily EC use s and 15 age- and sex-ma ched con ols), he e we e
no significan al e a ions in lung unc ion o eNO. Fu he mo e, HRCT
scans e ealed no significan s uc u al abno mali ies in he lungs o e
an a e age obse a ion pe iod o 3.5 yea s.
The isk o bias o he included s udies is epo ed in Supplemen a y
Table S3. The obse a ional s udy design p e en s in e ing causali y,
despi e he longi udinal da a.
29,31,32 , 34–37,39,40
Mo eo e , exposu e and
ou come da a om he na ional su eys a e sel - epo ed, po en ially in-
oducing ecall bias and epo ing e o s.
29,31,32 , 34–37,39,40
Al hough in
gene al, s udies epo ed analysis adjus ing o some key con ounde s,
o he con ounde s we e no assessed (eg alle gies, flu, exposu e o ai
pollu ion, amily his o y o as hma, physical ac i i y, o o he en i-
onmen al exposu es). Addi ionally, he small sample sizes o ne e -
smoking EC use s limi ed he s a is ical powe needed o de ec po en-
ial associa ions. The ollow-up ime o ou come occu ence was judged
’unclea ’ in all s udies, as sus ained changes in espi a o y unc ion may
ake longe o de elop and mani es . Finally, s a egies o manage miss-
ing da a we e epo ed in only wo s udies.
32 , 35
Discussion
We pe o med a sys ema ic e iew o 10 p ospec i e s udies on
EC use and espi a o y ou comes (including sel - epo ed diagnoses,
symp oms and lung unc ion) among adolescen s and adul s who ne e
smoked ciga e es. The majo i y analysed da a om he US PATH s udy.
O e all, none o he ou s udies examining sel - epo ed espi a o y di-
agnoses ound a s a is ically significan associa ion wi h baseline EC use
among ne e -smoking indi iduals. Collec i ely, he six s udies on espi-
a o y symp oms did no show a significan associa ion wi h EC use and
se e e espi a o y symp oms; he e was some enuous e idence o an as-
socia ion wi h coughing and wheezing symp oms, hough his was sen-
si i e o model specifica ions (eg he exac cu -off o espi a o y symp-
oms; how o he combus ible obacco use was handled). One s udy on
lung unc ion es s showed no significan abno mali ies o pa hological
findings among EC use s who ne e smoked.
33
Below, we discuss impo an conside a ions in in e p e ing his e -
idence. We e e eade s o ou mo e ex ensi e p io na a i e e iew
and c i ical app aisal
21 (which included many o he cu en eligible
s udies) o a de ailed discussion o s eng hs and weaknesses, bu sum-
ma ise common hemes he e. Fi s , knowledge o he iming o exposu e
and ou come is necessa y o examine he possible causal ela ionship be-
ween EC use and espi a o y disease. While p ospec i e s udy can en-
su e he necessa y empo al sequence, p ospec i e s udy design alone
is no sufficien , especially in he case o ch onic espi a o y condi ions
ha may ha e p eceded EC use. Fo example, as hma is o en diagnosed
in childhood.
41
This limi a ion applies o To e al
37
(as pa icipan s who
al eady had as hma we e included) and Sanchez-Rome o e al
34
( hough
his analysis epo ed a non-significan associa ion ega dless). Fo u-
na ely, se e al o he s udies employed a s onge design, by omi ing
pa icipan s who had p e-exis ing espi a o y condi ions a baseline ha
could ha e explained espi a o y ou comes (Ka ey e al ,
40
Pa el e al ,
31
Pe ez e al,
32
Sa gen e al ,
35
S e ens e al ,
36
Xie e al ,
39
and Reddy e
al ,
42
which was no included in his e iew (see Supplemen a y Table
S1 and ou na a i e e iew
21
).
Ano he impo an limi a ion is emaining s a is ical bias om o he
sou ces o con ounding. In pa icula , mos s udies in his e iew did
no accoun o use o o he combus ible obacco p oduc . An excep ion
was a supplemen a y analysis by Pe ez e al :
32 a e u he emo ing
pa icipan s who used o he obacco p oduc s a baseline, he e was
no emaining associa ion be ween exclusi e EC use and as hma ou -
comes. O he unaccoun ed- o con ounding ac o s (eg en i onmen al
pollu an s) could u he a enua e associa ions.
Two s udies, To e al
37
and Sa gen e al ,
35
we e e ained he e despi e
no es ic ing hei analysis o ne e -smoking indi iduals, as he anal-
ysis allowed an es ima ion o EC’s possible effec among ne e -smoking
indi iduals (ie by adjus ing o smoking s a us using ne e -smoking as
he e e ence g oup); howe e , he e may be emaining bias by he in-
clusion o o me ly and cu en ly smoking pa icipan s.
21
Ano he limi a ion o exis ing e idence is small sample size due o
he low p e alence o EC use among ne e -smoking indi iduals.
16–18
Fo example, Polosa e al
33
, he one s udy iden ified in ou e iew ha
examined objec i e es s o lung unc ion, had a small sample size (n = 9
ne e -smoking adul s who used ECs). Simila ly, Kenkel e al
29 iden i-
fied only 12 such cases –none o which had espi a o y condi ions a
ollow-up – p e en ing s a is ical analysis o his g oup. Thus, he ab-
sence o e idence in his e iew should no be in e p e ed as e idence
o absence.
O he limi a ions o he s udies e iewed include a limi ed ollow-up
pe iod (1–5 yea s), o e - eliance on US da a (especially om PATH),
and defining ’EC use’ o e ly b oadly a he han using de ailed use pa -
e ns ha would allow an assessmen o cumula i e, ch onic use (see
Selya e al
43
). Fu u e esea ch should p io i ise longe ollow-ups, inde-
penden samples (especially non-US samples), and collec mo e de ailed
EC use pa e ns ha would be equi ed o e alua e a dose– esponse e-
la ionship.
Conclusions
This sys ema ic e iew and quali a i e syn hesis indica e ha EC
use among ne e -smoking indi iduals may pose mild isks o coughing
and wheezing, bu e idence is lacking o mo e significan o clinically
meaning ul espi a o y symp oms o ha ms o e he sho o medium
e m ( ∼1–5 yea s). Due o he limi a ions o a ailable da a, ongoing e-
sea ch on long- e m and hea y EC use is wa an ed o moni o espi a-
o y isks, especially esea ch using la ge and longe - e m p ospec i e
s udies,and assessing mo e de ailed EC use pa e ns.
Funding
The au ho (s) ecei ed no financial suppo o his a icle wi h he
excep ion o he con ibu ion om Depa men o Clinical and Expe i-
men al Medicine a he Uni e si y o Ca ania o co e publica ion ees
( UPB 6C725202048/2024 ).
Decla a ion o compe ing in e es
RP is ull enu ed p o esso o In e nal Medicine a he Uni e -
si y o Ca ania (I aly) and Medical Di ec o o he Ins i u e o In e -
nal Medicine a he same Uni e si y. He has ecei ed he ollowing
EU and go e nmen al compe i i e g an s: U-BIOPRED, AIR-PROM, In-
eg al Rheuma ology & Immunology Specialis s Ne wo k (IRIS), Min-
is e o dell’Uni e si à e della Rice ca (MUR) PNRR 3277/2021, PNRR
341/2022, and PNRR 411/2021 unded by Nex Gene a ionEU o he Eu-
opean Commission. He has also ecei ed in es iga o -ini ia ed g an s
om Founda ion o a Smoke-F ee Wo ld, Pfize , GlaxoSmi hKline,
CV The apeu ics, Neu oSea ch A/S, Sandoz, Me k Sha p & Dohme,
Boeh inge Ingelheim, No a is, A bi G oup S l., Duska The apeu ics,
and Fo es Labo a o ies. He is he ounde o he Cen e o Tobacco
P e en ion and T ea men (CPCT) and o he Cen e o Excellence o
he Accele a ion o Ha m Reduc ion a Ca ania Uni e si y. He has e-
cei ed consul ancy ees om Pfize , Boeh inge Ingelheim, Duska The -
apeu ics, Fo es Labo a o ies, CV The apeu ics, Se mo Inc., GRG Heal h,
Cla i a e Analy ics, Guidepoin Expe Ne wo k, and GLG G oup. He
ecei es ex books oyal ies om Else ie and EDRA. He is also Chai
o he Eu opean Technical Commi ee o S anda diza ion on “Re-
qui emen s and es me hods o emissions o elec onic ciga e es ”
(CEN/TC 437; WG4) and scien ific ad iso o he non-p ofi Founda ion
RIDE2Med. A ielle Selya epo s a ela ionship wi h Pinney Associa es
Pi sbu gh ha includes: employmen , nonfinancial suppo , and a el
eimbu semen . A ielle Selya epo s a ela ionship wi h Juul Labs Inc
9
