Jou nal o In ec ion 84 (2022) 329–336
Con en s lis s a ailable a ScienceDi ec
Jou nal o In ec ion
jou nal homepage: www.else ie .com/loca e/jin
Nasopha yngeal Mic obio a as an ea ly se e i y bioma ke in
COVID-19 hospi alised pa ien s
Ma ia Paz Ven e o
a , 1
, Osca Mo eno-Pe ez
b , c , 1
, Ca men Molina-Pa dines
a , ∗,
And eu Pay u í-Galla
d , e
, Vicen e Boix
, c
, Isabel Esc ibano
a
, I ene Galan
g
,
Pila González-delaAleja
, Ma io López-Pé ez
h , a
, Rosa io Sánchez-Ma ínez
i
,
Espe anza Me ino
, 2
, Juan Ca los Rod íguez
a , h , 2
a
Mic obiology Se ice, Alican e Gene al Uni e si y Hospi al - Alican e Ins i u e o Heal h and Biomedical Resea ch (ISABIAL), Alican e, Spain
b
Endoc inology and Nu i ion depa men , Alican e Gene al Uni e si y Hospi al - Alican e Ins i u e o Heal h and Biomedical Resea ch (ISABIAL), Alican e,
Spain
c
Clinical Medicine depa men , Miguel He nández Uni e si y, Elche, Spain
d
Sequen ia Bio ech, Ca e Com e d’U gell 240, 08036 Ba celona, Spain
e
Open Uni e si y o Ca alonia (UOC), Rambla del Poblenou, 156, 08018 Ba celona, Spain
Uni o In ec ious Diseases, Alican e Gene al Uni e si y Hospi al - Alican e Ins i u e o Heal h and Biomedical Resea ch (ISABIAL), Alican e, Spain
g
Pneumology depa men , Alican e Gene al Uni e si y Hospi al - Alican e Ins i u e o Heal h and Biomedical Resea ch (ISABIAL), Alican e, Spain
h
E olu iona y Genomics G oup, Di isión de Mic obiología, Uni e sidad Miguel He nández, Spain
i
In e nal Medicine depa men , Alican e Gene al Uni e si y Hospi al - Alican e Ins i u e o Sani a y and Biomedical Resea ch (ISABIAL), Alican e, Spain
a i c l e i n o
A icle his o y:
Accep ed 21 Decembe 2021
A ailable online 25 Decembe 2021
Keywo ds:
Mic obio a
COVID-19
SARS-COV-2
Se e i y
Bioma ke
P ognosis
s u m m a y
This s udy aimed o analyse he di e si y and axonomic composi ion o he nasopha yngeal mic obio a,
o de e mine i s associa ion wi h COVID-19 clinical ou come. To s udy he mic obio a, we u ilized 16S
RNA sequencing o 177 samples ha came om a e ospec i e coho o COVID-19 hospi alized pa ien s.
Raw sequences we e p ocessed by QIIME2. The associa ions be ween mic obio a, in asi e mechanical en-
ila ion (IMV), and all-cause mo ali y we e analysed by mul iple logis ic eg ession, adjus ed o age,
gende , and como bidi y. The mic obio a αdi e si y indexes we e lowe in pa ien s wi h a a al ou come,
whe eas he βdi e si y analysis showed a significan clus e ing in hese pa ien s. A e mul i a ia e ad-
jus men , he p esence o Selenomonas spp., Fili ac o spp., Ac inobacillus spp., o Ch oococcidiopsis spp.,
was associa ed wi h a educ ion o mo e han 90% o IMV. Highe di e si y and he p esence o ce ain
gene a in he nasopha yngeal mic obio a seem o be ea ly bioma ke s o a a ou able clinical e olu ion
in hospi alized COVID-19 pa ien s.
©2022 The B i ish In ec ion Associa ion. Published by Else ie L d. All igh s ese ed.
In oduc ion
In his ime o pandemic finding ea ly p ognos ic ma ke s o
COVID-19 se e i y is o u mos impo ance [ 1 , 2 ]. I is known ha
poo ou comes ela ed o COVID-19 a e no only a consequence o
he i al in ec ion, bu a e also ela ed o an abe an hos immune
esponse, including he as elease o cy okines by he immune
sys em, leading o uncon olled inflamma ion and mul i-o gan ail-
u e [3] .
∗Co esponding au ho .
E-mail add ess: [email p o ec ed] (C. Molina-Pa dines).
1 These au ho s con ibu ed o he manusc ip equally and sha e he fi s au ho -
ship
2 Espe anza Me ino and Juan Ca los Rod íguez a e Join Senio Au ho s
Se e al isk o p ognos ic ac o s, such as gene ic ac o s, co-
mo bidi ies, age, sex, and geog aphical loca ion, ha e been associ-
a ed wi h COVID-19 se e i y [ 2 , 4 , 5 ]. Taken oge he , hese cha ac-
e is ics could ha e a de e mining ole in p omo ing immune e-
sponses and p e en ing an excessi e an i- i al immune eac ion.
Mic obio a may be ela ed o o influence he na u al his o y
o ce ain in ec ious diseases [6] . Fo example, in Clos idioides di -
ficile in ec ion, a lowe di e si y o mic obio a and a dec ease in
se e al amilies a e associa ed wi h he incidence and clinical e o-
lu ion o he disease [7] . Likewise, he espi a o y mic obio a has
also been co ela ed wi h he clinical e olu ion o ch onic espi a-
o y diseases [8] and espi a o y i al in ec ions [9] .
Rega ding mic obio a and COVID-19 pa hology, many published
s udies ha e ocused on he di e ences be ween COVID-19 and
non-COVID-19 pa ien s, sugges ing a possible ole o he gu o
h ps://doi.o g/10.1016/j.jin .2021.12.030
0163-4453/© 2022 The B i ish In ec ion Associa ion. Published by Else ie L d. All igh s ese ed.
M.P. Ven e o, O. Mo eno-Pe ez, C. Molina-Pa dines e al. Jou nal o In ec ion 84 (2022) 329–336
espi a o y mic obio a in suscep ibili y o SARS-CoV-2 in ec ion
[ 10 , 11 ]. Addi ionally, some s udies ha e shown a ela ionship be-
ween he composi ion o he gu and espi a o y mic obio a
and disease se e i y [12] . This ela ionship appea s o be mainly
based on he capaci y o he mic obio a o modula e he im-
mune esponse [ 13 , 14 ], h ough modifica ion o he gu -lung axis
[ 12 , 15 , 16 ], and o al e he exp ession o angio ensin-con e ing
enzyme 2 (ACE2) ecep o s, which a e used by SARS-CoV-2 o en-
e hos cells [ 17 , 18 ].
The a ailable e idence sugges s a po en ial ole o mic obio a in
suscep ibili y o SARS-CoV-2 in ec ion and COVID-19 se e i y, bu
longi udinal s udies e alua ing he mic obio a as a p ognos ic ac-
o o se e i y o disease p og ession a e lacking [19–22] . The da a
ega ding he associa ion be ween nasopha yngeal mic obio a ea-
u es and disease se e i y a e sca ce and limi ed in e ms o show-
ing a dec ease in αdi e si y o iden i ying specific gene a wi h
ele ance o c i ical illness [ 23 , 24 ]. Since he sampling o his loca-
ion is e y accessible, wi h he nasopha yngeal aspi a e swab di-
agnos ic confi ma ion p ocedu e able o ob ain his in o ma ion, i
should be a p io i y o add ess he ela ionship be ween nasopha-
yngeal mic obio a and COVID-19 ou comes. In his line, i would
be in e es ed eplica e his esul s in sali a samples, since sali a
has demons a ed a high simila i y in e ms o SARS-CoV-2 de ec-
ion [25–27] , and i is easie and less unpleasan o be ob ained.
This s udy aimed o analyse he nasopha yngeal mic obio a
om hospi alised COVID-19 pa ien s, o de e mine he ela ionship
be ween he mic obio a and SARS-CoV-2 in ec ion clinical ou -
comes and o iden i y ea u es o gene a ha could be used as
se e i y p ognos ic ma ke s.
Ma e ials and me hods
Pa ien s and s udy design and se ing
A e ospec i e coho o adul pa ien s wi h COVID-19, hospi-
alised in a e ia y cen e (Alican e Uni e si y Gene al Hospi al,
Spain) om Feb ua y 27 h 2020 o Janua y 22nd 2021, was s udied.
SARS-CoV-2 in ec ion was confi med by he RT-PCR-COBAS 6800
Sys em (Roche Molecula Sys ems, B anchbu g, NJ, Uni ed S a es).
A hospi al admission one nasopha yngeal specimen pe pa ien
was ob ained, s o ed a −80 °C and la e analysed.
O he 1526 pa ien s hospi alised in he s udy pe iod, nasopha-
yngeal samples om 324 pa ien s we e andomly p ocessed and
p ese ed. Due o he a ailable economic esou ces, six y pe cen
o he samples we e andomly sampled o p ocessing; 17 samples
did no co espond o he fi s PCR sample, so hey we e disca ded.
Finally, 177 pa ien s we e included in he s udy (see supplemen-
a y ma e ial, Figu e S1).
Va iables and da a collec ion
The clinical ea u es, como bidi y, labo a o y and adiological
es s, p esc ibed he apies, and ou come du ing he acu e phase o
he in ec ion by SARS-CoV-2 we e ex ac ed om he digi al med-
ical eco d.
The main explana o y a iables o he analysis we e he mic o-
bio a di e si y, measu ed by he αand βdi e si y indexes, and he
axonomic composi ion, exp essed by he di e en ially ep esen ed
gene a.
P ima y Ou comes: In asi e mechanical en ila ion (IMV) and
all-cause mo ali y.
DNA isola ion and mic obio a amplicon nex -gene a ion sequencing
(NGS)
The nasopha yngeal samples ozen a –80 °C we e used o
DNA isola ion wi h he QIAamp MiniElu e Vi us Spin Ki (Quia-
gen, Hilden, Ge many), ollowing he p o ocol ecommended by he
manu ac u e . The DNA ob ained was quan ified wi h a Qubi 4
Fluo ome e , using a Qubi dsDNA HS Assay Ki (The moFishe Sci-
en ific, Massachuse s, Uni ed S a es). The mic obio a amplicon se-
quencing was pe o med ollowing he p o ocol o he 16S Me age-
nomics Sequencing Lib a y P epa a ion ecommended by Illumina.
The V3 and V4 egion om 16S RNA gene we e amplified by PCR,
and hen he agmen s ob ained we e sequenced in he MiSeq
sys em wi h V3 eagen s (600 cycle, 2 ×300 bp).
Bioin o ma ic analyses
The aw eads ob ained om he NGS we e analysed using QII-
IME2 (2021.2 e sion) [28] . The denoising was pe o med wi h he
plugin DADA2 and o a oid con amina ion and alse posi i es a
BLAST agains he da abase o human genome o NCBI was pe -
o med, as well as single ons we e emo ed. The axonomy was
assigned using he SILVA Da abase (Release 132) [29] . Rega ding
he mic obio a analyses, he Shannon, Pielou, and Simpson indexes
we e calcula ed o s udy he αdi e si y, and he UniF ac weigh ed
dis ance plus PCoA we e pe o med o analyse he βdi e si y.
The gene a ha we e di e en ially ep esen ed be ween se e i y
g oups (main ou comes p esen o no ) we e de e mined using he
R package DESeq2 (4.1.0 e sion) [30] . The linea model ob ained
by DESeq2 was adjus ed by he p esc ip ion o an ibio ic ea men
3 mon hs ea lie .
S a is ical analysis
Ca ego ical and con inuous a iables a e gi en as equencies
(pe cen ages) and as he median (in e qua ile ange), espec i ely.
Pa ien s o he global coho ha we e included and excluded we e
compa ed by Mann-Whi ney’s U, chi-squa ed, and Fishe ’s exac
es s. Cumula i e incidences o ou comes (95% confidence in e als
(95%CI)) we e egis e ed. The final da e o ollow-up was Ma ch
1, 2021, unless censo ed. The di e ences be ween g oups in he β
di e si y we e assessed using he PERMANOVA es . Associa ions
we e e alua ed by a chi-squa ed es . Mul iple logis ic eg ession
models adjus ed o age, gende , and como bidi y we e buil o
e alua e he associa ion be ween mic obio a di e si y indexes o
he di e en ially ep esen ed genus (ob ained by DESeq2) wi h he
p ima y ou comes, and he odds a ios (OR) wi h he 95%CI we e
es ima ed. IBM SPSS S a is ics 25 (A monk, NY) was used o he
analyses. P < 0.050 defined s a is ical significance.
E hics s a emen and da a a ailabili y
This p ojec was pe o med in he Clinical and Biomedical Re-
sea ch Ins i u e o Alican e (ISABIAL), unde he w i en app o al
o he E hics Commi ee o Clinical Resea ch wi h D ugs (in Span-
ish, CEIm) o he Gene al Uni e si y Hospi al o Alican e (Re CEIm
app o al: PI2020–052).
The aw da a om he sequencing a e a ailable in he Na ional
cen e o Bio echnology In o ma ion Da abase (NCBI), unde he
Biop ojec accession numbe PRJNA754005.
Resul s
Pa icipans and desc ip i e da a
A o al o 177 pa ien s we e included in he s udy. The s udy
popula ion and he global coho o 1526 pa ien s hospi alised
330
M.P. Ven e o, O. Mo eno-Pe ez, C. Molina-Pa dines e al. Jou nal o In ec ion 84 (2022) 329–336
Table 1
Demog aphic cha ac e is ics, como bidi ies, clinical p esen a ion, and clinical ou comes.
∗.
Popula ion[ n = 177]
Demog aphics
Age, median (IQR), yea s 68 (52–80)
Age ≥65 yea s old,% (N) 55.9 (99/177)
Males,% (N) 57.6 (102/177)
Nosocomial,% (N) 1.7 (3/177)
Long- e m ca e esiden ,% (N) 4 (7/177)
Heal h p o essional,% (N) 4 (7/177)
Wa esFi s (1.02.2020 - 31.05.2020),% (N)Second (1.06.2020 - 15.12.2020),% (N)Thi d (16.12.2020 - 31.03.2021),% (N) 54.2 (96/177)31.1 (55/177)14.7 (26/177)
An ibio ic he apy in he p e ious 3 mon hs 28.8 (51/177)
Como bidi ies
Hype ension,% (N) 55.9 (99/177)
Diabe es,% (N) 26.6 (47/177)
Cu en o o me Smoke ,% (N) 20.6 (70/177)
Obesi y,% (N) 39.7 (56/141)
Ch onic espi a o y disease,% (N) 21.6 (38/177)
Immunosupp ession,% (N) 4 (7/177)
Cha lson como bidi y index, median (IQR) 3 (1–6)
Cha lson index ≥3,% (N) 59.3% (105/177)
10-yea s expec ed su i al
a 53.3 (1.6–90.1)
Clinical P esen a ion
Median ime (IQR) om symp om o hospi aliza ion, days
b 6 (3–7)
Fe e ,% (N) 67.2 (119/177)
Cough,% (N) 26.0 (46/177)
Dyspnoea,% (N) 57.6 (102/177)
Dia hoea,% (N) 25 (447,177)
Con usion,% (N) 9.6 (17/177)
Fa igue,% (N) 41.0 (71/173)
Myalgias-a h algias,% (N) 30.1 (52/173)
Anosmia-dysgeusia,% (N) 6.9 (12/173)
Ini ial Assessmen
Oxime y < 94% a oom ai ,% (N) 43.7% (73/167)
PaO2:FiO2, median (IQR) 332 (272–404)
Respi a o y a e, b ea hs/min, median (IQR) 18 (16–24)
Sys olic BP, mmHg, median (IQR) 130 (118–145)
Dias olic BP, mmHg,median (IQR) 78 (68–89)
Tempe a u e, °C,median (IQR) 36.9 (36.3–37.7)
Hea a e, bea s/min, median (IQR) 92 (81–102)
eGFR, ml/min/m
2
, median (IQR) 73 (47–90)
Lymphocy es, pe mm
3
, median (IQR) 910 (700–1370)
Lymphopenia,% (N) 44.3 (78/176)
C- eac i e p o ein > 10 mg/dL,% (N) 33.1 (55/175)
P ocalci onin > 0.5 ng/mL,% (N) 12.4 (20/161)
Fe i in > 500 mg/L,% (N) 59.8 (98/164)
Lac a e dehyd ogenase > 250 U/L,% (N) 33.9 (53/156)
D-dime s > 1 mg/mL,% (N) 33.1 (53/160)
In e leukin 6 ≥10 pg/mL,% (N) 77.7 (101/130)
T oponin T > 14 ng/L,% (N) 49.4 (77/176)
B ain na iu e ic pep ide > 125 pg/mL,% (N) 53.5 (84/157)
Po assium mmol/L, median (IQR) 4.1 (3.8–4.4)
Pneumonia on X- ays,% (N) 89.2 (157/176)
Opaci ies > 50% o lung su ace on X- ays,% (N) 21.5 (38/177)
T ea men
Co icos e oids,% (N) 46.3% (82/177)
Remdesi i ,% (N) 3.9% (7/177)
Tocilizumab,% (N) 23.7% (42/177)
Ou comes
Non-in asi e mechanical en ila ion equi emen ,% (N) 23.1 (41/177)
In asi e mechanical en ila ion equi emen ,% (N) 11.3 (20/177)
Mo ali y,% (N) 17.5 (31/177)
∗Da a shown as%, median (in e qua ile ange, IQR), unless specified o he wise. In bold, s a is ically significan di e ences. Pe cen ages may no o al 100 because o
ounding.
a 10-yea s expec ed su i al de i ed om Cha lson como bidi y index sco e.
b Days o symp oms be o e admission. OR: odds a io, 95%CI: 95% confidence in e al.
while he s udy las ed we e simila in age, gende , como bidi ies,
ex en o infil a es on ches adiog aph, dexame hasone use, du-
a ion o hospi aliza ion, and ou comes: IMV and mo ali y ( p >
0.05).
Table 1 shows he gene al cha ac e is ics o he s udy popula-
ion and he main ea u es o he COVID-19 acu e phase in ec ion
and i s clinical e olu ion. The pa ien s had a median age o 68.0
yea s (IQR) (52.0–80.0); 57.6% we e males and 59.3% had a Cha l-
son como bidi y index ≥3. They we e assessed in he eme gency
depa men a e a median o 6 [3–7] days o symp oms, and 89.2%
had pneumonia. Fi y-one pa ien s (28.8%) had ecei ed an ibio ic
he apy in he 3 mon hs p io o hei hospi al admission, o a
median o 5 [2–6] days. The mo ali y a e was 17.5% (95%CI, 12.6–
23.7) (31/177), and 11.3% (95%CI, 7.4–16.8) (20/177) equi ed IMV.
331
M.P. Ven e o, O. Mo eno-Pe ez, C. Molina-Pa dines e al. Jou nal o In ec ion 84 (2022) 329–336
Fig. 1. Di e si y analysis: Boxplo s ob ained o he Shannon index (A), Pielou index (B), and Simpson index (C). PCoA (p incipal coo dina es analysis) o he βdi e si y
dis ibu ion along he samples (D).
Di e si y analysis and ou comes
The αdi e si y indexes we e lowe in pa ien s wi h a a al
ou come: Shannon 3.59[2.86–4.42] s. 4.39[3.12–5.14], p = 0.014;
Pielou 0.58[0.50–0.67] s. 0.71[0.55–0.79], p = 0.007; and Simp-
son index 0.80[0.62–0.88] s. 0.89[0.76–0.94], p = 0.018 ( Figs 1 A,
1 B, and 1 C). The p o ec i e e ec o a g ea e mic obio a di e si y
pe sis ed o he Shannon (adjus ed OR (aOR) 0.654 [95%CI 0.448–
0.956], p = 0.028) and Pielou indexes (aOR 0.055[95%CI 0.003–
0.823], p = 0.036) a e adjus men o age, gende , and como -
bidi ies. The βdi e si y analysis showed a significan clus e ing
( p = 0.014), g ouping oge he he a al ou come pa ien s ( Fig 1 D).
In he case o IMV, nei he he αdi e si y indexes no βdi e si y
analyses showed any significan di e ences.
Taxonomic analysis and ou comes
S ep ococcus spp. (14.14%), S aphylococcus spp. (12.12%), and
Co ynebac e ium spp. (9.11%) we e he gene a mo e abundan in
COVID-19 pa ien s, wi hou significan di e ences be ween pa-
ien s wi h IMV o a a al ou come (see supplemen a y ma e ial,
Figu e S2, Table S1). By g oup, he e we e 34.20% (483/1412) axa
sha ed be ween IMV/non-IMV subpopula ions, 4.67% (66/1412)
axa exclusi ely ound in IMV pa ien s, and 61.12% (863/1412) axa
only de ec ed in non-IMV pa ien s ( Fig 2 A).
Rega ding a al ou come, he esul s we e simila . The sha ed
axa comp ised 41.57% (587/1412), axa exclusi ely ound in he
exi us subpopula ion we e 6.8% (96/1412), and in su i o s 51.2%
(729/1412) ( Fig 2 B).
332
M.P. Ven e o, O. Mo eno-Pe ez, C. Molina-Pa dines e al. Jou nal o In ec ion 84 (2022) 329–336
Fig. 2. Taxonomic analysis: Venn diag ams o IMV (A), and a al ou come (B), and ela i e abundances o di e en ial gene a o IMV (C), and a al ou come subpopula ions
(D). Rela i e abundances a e shown in loga i hmic scale. IMV: in asi e mechanical en ila ion, FO: Fa al ou come.
Di e en ly ep esen ed gene a and ou comes
This s udy was pe o med o iden i y di e en ial gene a be-
ween he subpopula ions wi h and wi hou specific ou comes. We
ound ha Selenomonas spp. (LogFC = 23.96; p < 0.0 0 01), Fili ac o
spp. (LogFC = 23.51; p < 0.0 0 01), Ac inobacillus spp. (LogFC = 24.86;
p < 0.0 0 01) and Ch oococcidiopsis spp. (LogFC = 22.31; p < 0.0 0 01)
we e significan ly mo e abundan in non-IMV pa ien s ( Fig 2 C,
Supplemen a y Table S2). The p esence o Selenomonas spp., Fili-
ac o spp. , Ac inobacillus spp. , o Ch oococcidiopsis spp. , was asso-
cia ed wi h a educed isk o IMV (OR 0.062 [95%CI 0.01–0.47],
p = 0.007). This p o ec i e associa ion pe sis ed a e adjus men
o he main con ounde s in he mul i a ia e model ( Fig 3 ).
Fo a al ou comes, Ac inobacillus spp. (LogFC = 24.30; p < 0.0 0 01),
Ci obac e spp. (LogFC = 25.21; p < 0.0 0 01), C au ococcus spp.
(LogFC = 22.77; p < 0.0 0 01), and Moheibac e spp. (LogFC = 22.7;
p < 0.0 0 01) we e significan ly mo e abundan in non-exi us pa ien s
( Fig 2 D, Supplemen a y Table S2). The p esence o Ac inobacillus
spp., Ci obac e spp. , C au ococcus spp. , o Moheibac e spp. , was
associa ed wi h a educed isk o a a al ou come (OR 0.309[95%CI
0.10–0.93], p = 0.037). This associa ion did no pe sis a e adjus -
men o he main con ounde s in he mul i a ia e model ( Fig 3 ).
Discussion
Recen ly, se e al s udies assessing he ela ionship be ween he
gu mic obiome and he se e i y o COVID-19 ha e been published
[ 31 , 32 ]. Howe e , o ou knowledge, his is he fi s s udy ha has
e alua ed nasopha yngeal mic obio a a he ime o admission as a
p ognosis bioma ke o se e i y o disease p og ession in he acu e
in ec ion phase o SARS-CoV-2, in a la ge coho o hospi alised
pa ien s wi h COVID-19. The assessmen showed a significan de-
c ease o all di e si y indexes s udied (Shannon, Pielou, and Simp-
son) in pa ien s wi h a final a al ou come, linking an ini ial low
mic obio a di e si y wi h COVID19 se e i y. The p esence o ou
specific gene a, Selenomonas spp., Fili ac o spp., Ac inobacillus spp.
o Ch oococcidiopsis spp., was associa ed wi h a educ ion o mo e
han 90% o IMV, ega dless o age, gende , o como bidi y. The
p esence o Ac inobacillus spp., Ci obac e spp. , C au ococcus spp.
o Moheibac e spp. was associa ed wi h a 70% educ ion in mo -
ali y, bu his ela ionship did no pe sis a e adjus men o he
main con ounde s.
The ela ionship be ween he mic obio a and COVID-19 is an ac-
i e and expanding field o esea ch. P e ious s udies ha e been o-
cused on he di e ences o he gu mic obio a be ween COVID-19
333
M.P. Ven e o, O. Mo eno-Pe ez, C. Molina-Pa dines e al. Jou nal o In ec ion 84 (2022) 329–336
Fig. 3. P edic o s o In asi e Mechanical Ven ila ion and In-Hospi al Dea h om Mul i a iable Logis ic-Reg ession Analysis. The 95% confidence in e als (CIs) o he odds
a ios ha e been adjus ed o mul iple es ing.
and non-COVID19 pa ien s, o i s co ela ion wi h se e i y inflam-
ma o y ma ke s [ 10 , 11 ]. Howe e , he e has been limi ed in es i-
ga ion in o he ela ionship be ween mic obial communi ies and
COVID-19 clinical ou come.
Rega ding COVID-19 and he gu mic obiome, Gu e al. [33] e-
po ed ha COVID-19 pa ien s had a lowe di e si y mic obio a
(Shannon and Chao1 index) han heal hy con ols; also, se e al mi-
c oo ganisms ( S ep ococcus spp. , Ro hia spp. , Veillonella spp. and
Ac inomyces spp. ) we e iden ified ha could be used as COVID-
19 bioma ke s. Acco ding o hese da a, Zuo e al. [34] , using he
B ay-Cu is dissimila i ies es , desc ibed al e a ions in he gu mi-
c obiome a he whole genome le el, since hei COVID-19 pa-
ien s we e mo e he e ogeneous han heal hy con ols. Yeoh e al.
[12] ound ha specific gene a, such as Bifidobac e ium adolescen-
is, Eubac e ium ec ale , and Faecalibac e ium p ausni zii , we e de-
ple ed in he COVID-19 coho when compa ed wi h non-COVID-
19 pa ien s, and we e nega i ely co ela ed wi h he inflamma o y
ma ke CXCL10. The same co ela ion was epo ed by Zou e al.
[34] . Likewise, Gou e al. [35] showed ha he Bac e oides genus,
and specifically B. o a us , was associa ed wi h inflamma o y cy-
okines such as IL-6, TNF- αand IFN- γ[35] . These deple ed species
in COVID-19 pa ien s a e known o play immunomodula o y oles
in he human gas oin es inal sys em [36] .
In e ms o he associa ion o he uppe espi a o y ac mi-
c obiome and SARS-COV-2 in ec ion, he s udies pe o med o da e
ha e included small coho s o pa ien s. B aun e al. [37] ( n = 33),
De Maio e al. [38] ( n = 40), and Liu e al. [39] ( n = 9) showed
no significan di e ences in he nasopha yngeal mic obial com-
muni y be ween COVID-19 and con ol pa ien s using α- βdi e -
si y and axonomic composi ional analysis. Whe eas Mos a a e al.
[40] ( n = 50) and Engen e al. [41] ( n = 19) epo ed a lowe αdi-
e si y (Chao1, Shannon, and Simpson indexes) in COVID-19 com-
pa ed o heal hy pa ien s, and bo h g oups showed significan dis-
simila i ies in βdi e si y. The e o e, he e is con o e sy ega ding
lung and nasopha yngeal mic obio a composi ion on SARS-CoV2
in ec ion.
Rega ding mic obio a and COVID-19 se e i y, Ma e al. [23] ex-
plo ed he o opha yngeal mic obiome in COVID-19 pa ien s
( n = 31) wi h a ious se e i ies (mild, mode a e, se e e, o c i -
ical) compa ed wi h flu pa ien s ( n = 29) and heal hy con ols
( n = 28) using high- h oughpu me agenomics. They showed ha
c i ical COVID-19 pa ien s p esen ed wi h a significan diminu ion
in αdi e si y (Shannon index), while nonc i ical pa ien s exhibi ed
no significan change om he no mal g oup.
The p esen wo k pionee ed he analysis o he nasopha yngeal
mic obio a (using 16S RNA gene sequencing), in a la ge coho
o hospi alised pa ien s wi h COVID-19, as a p ognosis bioma ke .
The lowe di e si y in pa ien s wi h a a al ou come is in ag ee-
men wi h he hypo hesis ha low mic obio a di e si y is associ-
a ed wi h he de elopmen o se e al pa hologies [ 42 , 43 ], and high
di e si y is associa ed wi h lowe se e i y [44] .
A s udy pe o med wi h 24 c i ically ill COVID-19 pa ien s
and 24 non-COVID-19 pa ien s wi h pneumonia [45] showed ax-
onomical di e ences be ween he lung mic obio a o COVID-19
and non-COVID-19 pa ien s. The cha ac e is ic mic oo ganisms o
COVID-19 pa ien s we e Pseudomonas alcaligenes, Sphingobac e ium
334
M.P. Ven e o, O. Mo eno-Pe ez, C. Molina-Pa dines e al. Jou nal o In ec ion 84 (2022) 329–336
spp ., Clos idium hi anonis and Acine obac e schindle i . While he
gene a ha cha ac e ised he lung mic obio a in he COVID-
19-nega i e pa ien s we e S ep ococcus spp., Haemophilus o Se-
lenomonas spp. Rega ding he uppe espi a o y ac mic obio a,
Ma e al. [23] ound inc eased a ios o Klebsiella sp., Acine obac e
sp., and Se a ia sp. we e co ela ed wi h bo h disease se e i y and
ele a ed sys emic inflamma ion ma ke s (neu ophil–lymphocy e
a io). Along he same lines, P e o ella spp. was also linked o
COVID-19 se e i y, which has been hypo hesised o sugges a pos-
sible ela ionship wi h he inflamma o y esponse [24] .
Ou axonomic analysis iden ified se e al mic oo ganisms, such
as Selenomonas, Fili ac o , Ac inobacillus , and Ch oococcidiopsis SAG
2023, ela ed o IMV, and C au ococcus, Ac inobacillus, Ci obac e
and Moheibac e ela ed o a a al ou come. Fu u e esea ch o de-
e mine hei oles in COVID-19 de elopmen and e olu ion is e-
qui ed.
Ou s udy has se e al limi a ions, his was an obse a ional, e -
ospec i e, single-cen e s udy, and collec ion o da a was no s an-
da dized in ad ance. The sample size and he absence o di e -
ences in he cha ac e is ics o he global coho o pa ien s admi -
ed o ou hospi al du ing he du a ion o he s udy ein o ce he
p esen da a. Mul iple ac o s can condi ion changes in mic obio a,
including he use o an ibio ics. None heless, he design o he s a-
is ical analysis adjus ed o he use o an ibio ic he apy in he 3
mon hs p io o he inclusion o he s udy, allowing us o limi
his bias. The exclusion o hese pa ien s om he s udy would
ha e g ea ly limi ed he ex e nal alidi y o ou esul s. Finally,
he 16S ibosomal RNA amplicon sequencing app oach o s udy he
mic obio a could in oduce bias in he ob ained da a because his
me hod does no allow he s udy o he whole mic obiome, bu
only he gene a amplified by PCR. Ne e heless, i is he mos com-
mon echnique o s udy mic obio a in clinical samples. Mo eo e ,
he mic obio a bioin o ma ics analysis has no been s anda dized
ye , which hampe ed compa ison in e p e a ions o ou esul s.
In summa y, he highe di e si y ound in pa ien s wi hou IMV
o a a al ou come, oge he wi h he p esence o ce ain gene a in
he nasopha yngeal mic obio a, seemed o be an ea ly bioma ke
o a a ou able clinical e olu ion in a coho o Medi e anean hos-
pi alised pa ien s wi h SARS-CoV-2 in ec ion. Ou findings ha e po-
en ial clinical ele ance due o he easibili y and low cos o de-
eloping apid molecula echniques o e alua e he di e si y and
de ec hese gene a a he ime o admission. These da a, aken o-
ge he wi h o he p ognos ic ma ke s al eady being implemen ed,
may allow iden i ying pa ien s wi h a good p ognosis (i.e., a 70–
90% educ ion in un a ou able clinical ou comes). Conside ing he
clinical significance o hese findings and he ease o hei appli-
ca ion in daily p ac ice, u he in es iga ion o confi m hese da a
could be e y ele an o imp o ing COVID-19 managemen .
Funding
This wo k was suppo ed by he “Heal h Ins i u e Ca los III”
[g an numbe COV20_00236 ] and by he “Alican e Ins i u e o
Heal h and Biomedical Resea ch (ISABIAL)” [g an numbe exp:
2020–0356 ].
Decla a ion o Compe ing In e es
The au ho s decla e no conflic o in e es .
Acknowledgemen s
No applicable. Thanks o he COVID19-ALC Resea ch G oup o
he Gene al Uni e si y Hospi al o Alican e.
Supplemen a y ma e ials
Supplemen a y ma e ial associa ed wi h his a icle can be
ound, in he online e sion, a doi:10.1016/j.jin .2021.12.030 .
Re e ences
[1]. Liu J, Liu S. The managemen o co ona i us disease 2019 (COVID-19). J Med
Vi ol 2020; 92 :1484–90. doi: 10.1002/jm .25965 .
[2]. And és M, Leon-Rami ez JM, Mo eno-Pe ez O, Sánchez-Payá J, Gayá I, Es e-
ban V, e al. Fa ali y and isk ea u es o p ognosis in COVID-19 acco ding o
he ca e app oach - a e ospec i e coho s udy. PLoS ONE 2021; 16 :e0248869.
doi: 10.1371/jou nal.pone.0248869 .
[3]. Tay MZ, Poh CM, Rénia L, MacA y PA, Ng LFP. The ini y o COVID-19: immu-
ni y, inflamma ion and in e en ion. Na Re Immunol 2020 206 2020; 20 :363–
74. doi: 10.1038/s41577- 020- 0311- 8 .
[4]. Mo eno-P O, Leon-Rami ez JM, Fue es-Kenneally L, Pe digue o M, And es M,
Ga cia-Na a o M, e al. Hypokalemia as a sensi i e bioma ke o disease
se e i y and he equi emen o in asi e mechanical en ila ion equi emen
in COVID-19 pneumonia: a case se ies o 306 Medi e anean pa ien s. In J
In ec Dis 2020; 100 :449–54. doi: 10.1016/J.IJID.2020.09.033 .
[5]. Dai CL, Ko nilo SA,
Rope RT, Cohen-Cline H, Jade K, Smi h B, e al. Cha -
ac e is ics and ac o s associa ed wi h co ona i us disease 2019 in ec ion,
hospi aliza ion, and mo ali y ac oss ace and e hnici y. Clin In ec Dis 2021.
doi: 10.1093/CID/CIAB154 .
[6]. Pflughoe KJ, Ve salo ic J. Human mic obiome in heal h and
disease. Annu Re Pa hol Mech Dis 2012; 7 :99–122. doi: 10.1146/
annu e - pa hol- 011811-132421 .
[7]. Be kell M, Mysa a M, Xa ie BB, an We kho en CH, Monsieu s P, Lam-
mens C, e al. Mic obio a-based ma ke s p edic i e o de elopmen o
Clos idioides difficile in ec ion. Na Commun 2021 121 2021; 12 :1–14. doi: 10.
1038/s41467- 021- 22302- 0 .
[8]. Budden KF, Shukla SD, Rehman SF, Bowe man KL, Keely S, Hugenhol z P,
e al. Func ional e ec s o he mic obio a in ch onic espi a o y disease. Lance
Respi Med 2019; 7 :907–20. doi: 10.1016/S2213-2600(18)30510-1 .
[9]. Hanada S, Pi zadeh M, Ca e KY, Deng JC. Respi a o y i al in ec ion-induced
mic obiome al e a ions and seconda y bac e ial pneumonia. F on Immunol
2018; 9 :2640. doi: 10.3389/FIMMU.2018.02640 .
[10]. Ma cialis MA, Ba danzellu F, Fanos VMic obio a and Co ona i us Disease
Which came fi s , he chicken o he egg? Clin In ec Dis 2020 2019. doi: 10.
1093/cid/ciaa965 .
[11]. Yamamo o S, Sai o M, Tamu a A, P awisuda D, Mizu ani T, Yo suyanagi H.
The human mic obiome and COVID-19: a sys ema ic e iew. PLoS ONE
2021; 16 :e0253293. doi: 10.1371/JOURNAL.PONE.0253293 .
[12]. Yeoh YK, Zuo T, Lui GCY, Zhang F, Liu Q, Li AYL, e al. Gu mic obio a composi-
ion eflec s disease se e i y and dys unc ional immune esponses in pa ien s
wi h COVID-19. Gu 2021; 70 :698–706. doi: 10.1136/gu jnl- 2020- 323020 .
[13]. Kha iwada S, Subedi A. Lung mic obiome and co ona i us disease 2019
(COVID-19): possible link and implica ions. Hum Mic obiome J 2020; 17 . doi: 10.
1016/j.humic.2020.10 0 073 .
[14]. an de Lelie D, Tagha i S. COVID-19 and he Gu Mic obiome: mo e han a
Gu Feeling. MSys ems 2020; 5 . doi:
10.1128/msys ems.00453-20 .
[15]. Din AU, Mazha M, Wasim M, Ahmad W, Bibi A, Hassan A, e al. SARS-CoV-
2 mic obiome dysbiosis linked diso de s and possible p obio ics ole. Biomed
Pha maco he 2021; 133 . doi: 10.1016/j.biopha.2020.110947 .
[16]. Ka
´
zmie czak-Siedlecka K., Vi ale E., Maka ewicz W. COVID-19 – gas oin es i-
nal and gu mic obio a- ela ed aspec s. n.d. 2021
[17]. Ahlawa S, Asha Sha ma KK. Immunological co-o dina ion be ween gu and
lungs in SARS-CoV-2 in ec ion. Vi us Res 2020; 286 :198103. doi: 10.1016/j.
i us es.2020.198103 .
[18]. Koes e ST, Li N, Lachance DM, Mo ella NM, Dey N. Va iabili y in diges i e and
espi a o y ac Ace2 exp ession is associa ed wi h he mic obiome. PLoS ONE
2021; 16 :e0248730. doi: 10.1371/jou nal.pone.0248730 .
[19]. He Y, Wang J, Li F, Shi Y. Main clinical ea u es o COVID-19 and po en ial
p ognos ic and he apeu ic alue o he mic obio a in SARS-CoV-2 in ec ions.
F on Mic obiol 2020; 11 :1302. doi: 10.3389/ micb.2020.01302 .
[20]. Fe ei a C, Viana SD, Reis F. Is gu mic obio a dysbiosis a p edic o o in-
c eased suscep ibili y o poo ou come o COVID-19 pa ien s? An upda e. Mi-
c oo ganisms 2021; 9 :1–12. doi: 10.3390/mic oo ganisms9010053 .
[21]. He nández-Te án A, Mejía-Nepomuceno F, He e a MT, Ba e o O, Ga cía E,
Cas illejos M, e al. Dysbiosis and s uc u al dis up ion o he espi a o y mi-
c obio a in COVID-19 pa ien s wi h se e e and a al ou comes. Sci Repo s 2021
111 2021; 11 :1–13. doi: 10.1038/s41598- 021- 00851- 0 .
[22]. Ren L., Wang Y., Zhong J., Li X., Xiao Y., Li J., e al. Dynamics
o he Uppe Respi a o y T ac Mic obio a and i s Associa ion wi h
Mo ali y in COVID-19. h ps://doi.o g/101164/Rccm202103-0814OC 2021.
h ps://doi.o g/10.1164/RCCM.202103-0814OC.
[23]. Ma S, Zhang F, Zhou F, Li H, Ge W, Gan R, e al. Me agenomic analysis e-
eals o opha yngeal mic obio a al e a ions in pa ien s wi h COVID-19. Signal
T ansduc Ta ge The 2021 61 2021; 6 :1–11. doi: 10.1038/s41392- 021- 00614- 3 .
[24]. Ven e o MP, Cuad a RRC, Vidal I, And ade BGN, Molina-Pa dines C, Ha o-
Mo eno JM, e al. Nasopha yngeal mic obial communi ies o pa ien s in ec ed
wi h SARS-CoV-2 ha de eloped COVID-19. F on Mic obiol 2021; 12 :637430.
doi: 10.3389/ micb.2021.637430 .
335
M.P. Ven e o, O. Mo eno-Pe ez, C. Molina-Pa dines e al. Jou nal o In ec ion 84 (2022) 329–336
[25]. Ca ouel F, Vale e M, Pe ie H, Bouscambe -Duchamp M, Dussa C,
T amini P, e al. Pe o mance o sel -collec ed sali a es ing compa ed
wi h nasopha yngeal swab es ing o he de ec ion o SARS-CoV-2. Vi uses
2021; 13 :895. doi: 10.3390/V13050895 .
[26]. Boe ge AC, Buckwal e S, Fe nholz EC, Janne o PJ, Binnicke MJ, Reed K,
e al. E alua ion o sel -collec ed mid u bina e nasal swabs and sali a o de-
ec ion o sa s-co -2 na. J Clin Mic obiol 2021; 59 . doi: 10.1128/JCM.00848-21/
ASSET/D740D9AD- B977- 428E- B618- 18135BB129A8/ASSETS/IMAGES/LARGE/
JCM.00848- 21- F002.JPG .
[27]. Alemany A, Milla -Ma inez P, Ouchi D, Co bacho-Monné M, Bo doy AE, Es e-
ban C, e al. Sel -collec ed mid-nasal swabs and sali a specimens, compa ed
wi h nasopha yngeal swabs, o SARS-CoV-2 de ec ion in mild COVID-19 pa-
ien s. J In ec 2021. doi: 10.1016/J.JINF.2021.09.012 .
[28]. Kuczynski J, S ombaugh J, Wal e s WA, González A, Capo aso JG, Knigh R. Us-
ing QIIME o analyze 16S RNA gene sequences om mic obial communi ies.
Cu P o oc Mic obiol 2012 Chap e 1:Uni 1E.5. doi: 10.1002/9780471729259.
mc01e05s27 .
[29]. Quas C, P uesse E, Yilmaz P, Ge ken J, Schwee T, Ya za P, e al. The SILVA i-
bosomal RNA gene da abase p ojec : imp o ed da a p ocessing and web-based
ools. Nucleic Acids Res 2012; 41 :D590–6. doi: 10.1093/na /gks1219 .
[30]. Lo e MI, Hube W, Ande s S. Mode a ed es ima ion o old change and dis-
pe sion o RNA-seq da a wi h DESeq2. Genome Biol 2014 1512 2014; 15 :1–21.
doi: 10.1186/S13059- 014- 0550- 8 .
[31]. Pa el P, Rope J. Gu mic obiome composi ion is associa ed wi h COVID-19 dis-
ease se e i y. Gas oen e ology 2021; 161 :722–4. doi: 10.1053/J.GASTRO.2021.05.
006 .
[32]. D D, A M. Gu mic obio a and Co id-19- possible link and implica ions. Vi us
Res 2020; 285 . doi: 10.1016/J.VIRUSRES.2020.198018 .
[33]. Gu S, Chen Y, Wu Z, Chen Y, Gao H, L L, e al. Al e a ions o he Gu Mic o-
bio a in Pa ien s Wi h Co ona i us Disease 2019 o H1N1 Influenza.
Clin In ec
Dis 2020; 71 :2669–78. doi: 10.1093/CID/CIAA709 .
[34]. Zuo T, Zhang F, Lui GCY, Yeoh YK, Li AYL, Zhan H, e al. Al e a ions
in gu mic obio a o pa ien s wi h COVID-19 du ing ime o hospi al-
iza ion. Gas oen e ology 2020; 159 :944–55 e8. doi: 10.1053/J.GASTRO.2020.
05.048 .
[35]. Gou W, Fu Y, Yue L, Chen G, Cai X, Shuai M, e al. Gu mic obio a, inflamma-
ion, and molecula signa u es o hos esponse o in ec ion. J Gene Genomics
2021. doi: 10.1016/J.JGG.2021.04.002 .
[36]. Nishida A, Inoue R, Ina omi O, Bamba S, Nai o Y, Andoh A. Gu mic obio a in
he pa hogenesis o inflamma o y bowel disease. Clin J Gas oen e ol 2018; 11 .
doi: 10.1007/S12328- 017- 0813- 5 .
[37]. B aun T, Hale i S, Hada R, E oni G, Glick Saa E, Kelle N, e al. SARS-CoV-2
does no ha e a s ong e ec on he nasopha yngeal mic obial composi ion.
Sci Rep 2021; 11 :8922.
doi: 10.1038/s41598- 021- 88536- 6 .
[38]. De Maio F., Pos e a o B., Ponziani F.R., Ca ani P., Gasba ini A., Sanguine i M.
Nasopha yngeal Mic obio a P ofiling o SARS-CoV-2 In ec ed Pa ien s n.d. 2021
h ps://doi.o g/10.1186/s12575-020-00131-7.
[39]. Liu J, Liu S, Zhang Z, Lee X, Wu W, Huang Z, e al. Associa ion be ween he na-
sopha yngeal mic obiome and me abolome in pa ien s wi h COVID-19. Syn h
Sys Bio echnol 2021; 6 :135–43. doi: 10.1016/J.SYNBIO.2021.06.002 .
[40]. Mos a a HH, Fissel JA, Fanelli B, Be gman Y, Gniazdowski V, Dadlani M, e al.
Me agenomic nex -gene a ion sequencing o nasopha yngeal specimens col-
lec ed om confi med and suspec co id-19 pa ien s. MBio 2020; 11 :1–13.
doi: 10.1128/MBIO.01969-20
.
[41]. Engen PA, Naqib A, Jennings C, G een SJ, Landay A, Kesha a zian A, e al. Na-
sopha yngeal Mic obio a in SARS-CoV-2 Posi i e and Nega i e Pa ien s. Biol
P oced Online 2021 231 2021; 23 :1–6. doi: 10.1186/S12575- 021- 00148- 6 .
[42]. Kalan a KL, Moazed F, Ch is enson SC, Wilson J, Deiss T, Belze A, e al.
Me agenomic compa ison o acheal aspi a e and mini-b onchial al eola
la age o assessmen o espi a o y mic obio a. Am J Physiol - Lung Cell Mol
Physiol 2019; 316 :L578–84. doi: 10.1152/ajplung.00476.2018 .
[43]. Sol ani S, Zake i A, Zandi M, Kesheh MM, Tabibzadeh A, Das anj M, e al.
The ole o bac e ial and ungal human espi a o y mic obio a in COVID-19
pa ien s. Biomed Res In 2021 2021. doi: 10.1155/2021/6670798 .
[44]. Ogun inola GA, Oyewale JO, Oshamika OO, Olasehinde GI. The human mic o-
biome and i s impac s on heal h. In J Mic obiol 2020; 2020 . doi: 10.1155/2020/
8045646 .
[45]. Gaibani P, Viciani E, Ba ole i M, Lewis RE, Tone i T, Lomba do D, e al. The
lowe espi a o y ac mic obiome o c i ically ill pa ien s wi h COVID-19. Sci
Rep 2021; 11 :10103. doi: 10.1038/s41598- 021- 89516- 6 .
336
