Mortality predictors and definition proposal for complicated coagulase-negative Staphylococcus bacteraemia: a multicentre prospective cohort study

Varisco, Benedetta; Martínez Pérez-Crespo, Pedro María; Retamar Gentil, Pilar; López Hernández, Inmaculada; Fariñas Álvarez, Maria Carmen; Fernández-Natal, Isabel; Rodríguez-Baño, Jesús; López-Cortés, Luis E.
doi:10.1016/j.cmi.2024.12.016
Mortality predictors and definition proposal for complicated coagulase-negative Staphylococcus bacteraemia: a multicentre prospective cohort study

Author: Varisco, Benedetta; Martínez Pérez-Crespo, Pedro María; Retamar Gentil, Pilar; López Hernández, Inmaculada; Fariñas Álvarez, Maria Carmen; Fernández-Natal, Isabel; Rodríguez-Baño, Jesús; López-Cortés, Luis E.
Publisher: Wiley-Blackwell; Elsevier Sci Ltd
Year: 2025
DOI: 10.1016/j.cmi.2024.12.016
Source: https://idus.us.es/bitstreams/daa89f63-488a-492b-aeaa-a68d5ff9b793/download
Fo Pee Re iew
Mo ali y p edic o s and de ini ion p oposal o complica ed
coagulase-nega i e S aphylococcus bac e emia. A
mul icen e p ospec i e coho s udy
Jou nal:
Clinical Mic obiology and In ec ion
Manusc ip ID
CLM-24-28675.R1
A icle Type:
O iginal A icle
Da e Submi ed by he
Au ho :
n/a
Comple e Lis o Au ho s:
Va isco, Benede a; ASST San i Paolo e Ca lo, Clinic o In ec ious and
T opical Diseases, Depa men o Heal h Sciences
Ma ínez Pé ez-C espo, Ped o Ma ía; Hospi al Uni e si a io de Valme,
Unidad de En e medades In ecciosas y Mic obiología
Re ama , Pila ; Hospi al Uni e si a io Vi gen Maca ena, Unidad Clínica de
En e medades In ecciosas y Mic obiología
López-He nández, Inmaculada; Hospi al Uni e si a io Vi gen Maca ena,
Unidad Clínica de En e medades In ecciosas y Mic obiología
Fa iñas, Ma ia Ca men; Uni e si y Hospi al Ma qués de Valdecilla,
In ec ious Diseases
Fe nández-Na al, Isabel; Complejo Asis encial Uni e si a io de León,
Clinical Mic obiology
Ma ía Te esa, Pé ez-Rod íguez; Complexo Hospi ala io Uni e si a io de
Vigo, Se icio de Medicina In e na
Goikoe xea, Josune; Hospi al Uni e si a io de C uces, En e medades
In ecciosas
Sánchez Cal o, Juan Manuel; Hospi al Uni e si a io de Je ez de la
F on e a, UGC En e medades In ecciosas y Mic obiología Clínica
Buzon-Ma in, Luis; Hospi al Uni e si a io de Bu gos, Medicina In e na
León , E a; Hospi al Uni e si a io Nues a Seño a de Valme, Unidad de
En e medades In ecciosas y Mic obiología
Vinuesa Ga cía, Da id; Hospi al Uni e si a io Clínico San Cecilio., Unidad
de Ges ión Clínica de En e medades In ecciosas.
Regue a, José; Regional Uni e si y Hospi al o Malaga, Depa men o
In ec ious Diseases
Bahamonde, Albe o; Hospi al El Bie zo, Depa men o In e nal Medicine.
Fe nandez, Jona han; Hospi al Uni e si a io Cen al de As u ias,
Mic obiology Uni
Rod íguez-Baño, Jesús; Hospi al Uni e si a io Vi gen Maca ena,
En e medades In ecciosas, Mic obiologa y Medicina P e en i a;
Uni e sidad de Se illa, Medicina
López-Co és, Luis Edua do ; Hospi al Uni e si a io Vi gen Maca ena,
Unidad Clínica de En e medades In ecciosas y Mic obiología; Uni e sidad
de Se illa, Facul ad de Medicina
Key Wo ds:
Coagulase-nega i e s aphylococci, Bloods eam in ec ions, Complica ed
bac e emia, Mo ali y p edic o s, Ca he e - ela ed bloods eam in ec ions
Clinical Mic obiology and In ec ion
Fo Pee Re iew
Abs ac :
Objec i e. To explo e a de ini ion o complica ed coagulase-nega i e
s aphylococci bloods eam in ec ions (CoNS BSI), and o iden i y
p edic o s o mo ali y.
Me hods. P ospec i e coho s udy conduc ed om Oc obe 2016 o
Ma ch 2017 in 26 Spanish hospi als. Complica ed CoNS BSI c i e ia
included lack o ea ly ca he e emo al in ca he e - ela ed cases, o eign
indwelling implan , pe sis en bac e emia, e e ≥72 hou s on ac i e
he apy, me as a ic in ec ion o deep-sea ed ocus and in ec i e
endoca di is. Independen p edic o s o 30-day mo ali y we e e alua ed
by Cox eg ession, and he impac o he de ini ion o complica ed
bac e emia assessed.
Resul s. O e all, 445 CoNS BSI cases we e included; ca he e - ela ed
in ec ions we e p edominan (336/445, 75.5%). Complica ed bac e emia
was iden i ied in 240/445 pa ien s (53.9%); 30-day mo ali y in
complica ed and uncomplica ed cases we e 53/240 (22.1%) and 24/205
(11.7%), espec i ely (p=0.004). P edic o s o 30-day mo ali y iden i ied
in he mul i a ia e analysis included age (HR 1.03, 95%CI 1.01-1.05),
ce eb o ascula disease (HR 2.58, 95%CI 1.45-4.58),
immunosupp essi e he apy (HR 2.16, 95%CI 1.22-3.84), SOFA sco e
(HR 1.09, 95%CI 1.03-1.16), and complica ed bac e emia (HR 2.14,
95%CI 1.29-3.53). A ca he e - ela ed sou ce o bac e emia was ound o
be p o ec i e (HR 0.49, 95%CI 0.30-0.80). When speci ic c i e ia o
de ine complica ed bac e emia we e included, e e ≥72h was associa ed
wi h inc eased isk o dea h (HR 2.52, 95%CI 1.52-4.17) and ea ly
ca he e emo al was p o ec i e (HR 0.47, 95%CI 0.26-0.83).
Conclusions. A high p opo ion o pa ien s p esen ed complica ed
bac e emia acco ding o he p oposed c i e ia; hese pa ien s had highe
haza ds o mo ali y. O he mo ali y p edic o s we e iden i ied. Fu he
s udies would be needed o alida e he p oposed c i e ia.
Page 1 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
Se ille, Decembe 2024
Dea P o . Paul,
Edi o , Clinical Mic obiology and In ec ion
Thank you o he cons uc i e eedback and de ailed sugges ions ega ding ou manusc ip ,
"Mo ali y p edic o s and de ini ion p oposal o complica ed coagulase-nega i e S aphylococcus
bac e emia. A mul icen e p ospec i e coho s udy" (manusc ip ID: CLM-24-28675). We
app ecia e he ime and e o ha bo h you and he e iewe s ha e dedica ed o his p ocess.
Based on you sugges ions, we ha e made ca e ul e isions o imp o e he cla i y and s eng h o
ou wo k.
The acked and clean copies o he e ised manusc ip a e submi ed he e o you
conside a ion.
Kind ega ds,
Luis Edua do López-Co és
On behal o all au ho s
Page 2 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
EDITOR COMMENTS TO THE AUTHOR
Commen (1): Among inclusion c i e ia “ he e we e sys emic symp oms like e e o chills” –
was his he ac ual c i e ion used o inclusion? As de ined, his c i e ion is non-speci ic, canno
be eplica ed and p obably leads o he e ogenei y in pa ien inclusion. I a mo e p ecise c i e ion
was used please de ine i in he manusc ip . I his was he c i e ion, we ha e li le con idence
ha he pa ien s in he coho has a clinically-signi ican CONS bac e emia.
Response: Thank you o he impo an eedback. To be b ie , we acknowledge ha he
c i e ia we e no desc ibed wi h su icien speci ici y. We ha e now cla i ied he speci ic
inclusion c i e ia in he manusc ip and Supplemen a y ma e ial, aligning hem wi h hose
ou lined in he s udy p o ocol.
Commen (2): “Da a we e collec ed ia an elec onic case epo ” – please de ine da a sou ces
and collec ion me hods. We e da a manually collec ed om elec onic pa ien s iles o pape
cha s? Was any au oma ed da a que y ool used?
Response: We ha e e ised he manusc ip o cla i y he da a collec ion me hods, which
now eads: "Da a we e collec ed p ospec i ely om ei he elec onic heal h eco ds o
pape cha s a each cen e acco ding o local a ailabili y. Da a o each en olled
pa ien we e en e ed in o a s anda dized online case epo o m.”
Commen (3): Unde “De ini ions and a iables” please de ine “empi ic he apy” – i s 24
hou s? 48 hou s” wha is he s a ing ime poin – blood cul u e collec ion ime? In me hods you
de ined App op ia e an imic obial he apy and in he esul s you e e o “ac i e” he apy –
please use consis en e minology.
Response: Thank you o you eedback. We ha e added he de ini ion o "empi ic
he apy," which is now s a ed as " he an imic obial egimen ini ia ed p io o he
iden i ica ion o he in ec ing mic oo ganism and i s suscep ibili y p o ile; since
G am/MALDI epo s we e ypically a ailable wi hin 24 hou s, he apy was usually
conside ed empi ical du ing his ime ame". Rega ding he e minology inconsis ency,
we app ecia e you obse a ion. We ind he e m "ac i e he apy" mo e accu a e han
"app op ia e he apy." As such, we ha e e ised he manusc ip o ensu e he consis en
use o "ac i e he apy" h oughou he sec ions.
Commen (4): Follow-up blood cul u es we e aken only o <30% o pa ien s by physicians’
disc e ion. CONS a e low i ulence pa hogens who do no necessa y cause ma ked sepsis signs,
Page 3 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
i is likely ha some o he pa ien s in whom ollow-up blood cul u es we e no aken
(p esumably because hey had no e e o o he sepsis signs) also had p olonged bac e emia. I
p opose lis ing he indica ions o aking ollow-up blood cul u es (p ospec i e c i e ia o
e oac i e easons om he da a) and e-naming he a iable “Posi i e ollow-up blood cul u es”
o “Posi i e ollow-up blood cul u es when indica ed” o simila .
Response: Thank you o highligh ing his impo an poin . Gi en he lack o guidelines
o ollow-up blood cul u es (FUBC) in CoNS bac e emia, ou obse a ional s udy
e lec s clinical p ac ice, wi h FUBC pe o med a he disc e ion o he ea ing physician.
We ecognize ha posi i e FUBC esul s may ep esen only a subse o ac ual pe sis en
bac e emia cases, as some pa ien s no es ed may ha e had unde ec ed posi i e esul s.
To add ess his, we ha e included his poin in he Limi a ions sec ion o he s udy.
In esponse o you sugges ion, we conduc ed a e ospec i e analysis o ac o s ha may
ha e in luenced he decision o pe o m FUBC. As shown in Supplemen a y Table S1,
pa ien s wi h endoca di is o ca diac/endo ascula de ices we e signi ican ly mo e likely
o unde go FUBC. We ha e summa ized hese indings in he Resul s sec ion. Al hough
we hypo hesize ha hese lis ed ac o s likely in luenced FUBC decisions, ou da ase
does no include in o ma ion on he speci ic clinical indica ions o each FUBC, as no ed
in he Limi a ions.
Rega ding he a iable name, please no e ha in Table 1 we include he a iable “Blood
cul u es” (now called “Follow-up blood cul u es pe o med”), and we s a e a Table oo
ha FUBC we e pe o med a he disc e ion o a ending physician, and e e o ou
analysis. The o he a iable is now called "Posi i e ollow-up blood cul u es among hose
pe o med."
Commen (5): The e iewe s we e c i ical o he ca he e ex ac ion measu e among he
measu es de ining complica ed bac e emia, since he e is a consensus ha ca he e e aining is
possible in he ini ial managemen o CONS CRABSI bac e emia. I ag ee ha since he e is no
good e idence o his ecommenda ion his emains a ques ion. Howe e , his a iable is
ele an only o he subg oup o pa ien s wi h a CVC. The a iable “ea ly (<48h) ca he e
emo al” in model 2 mixes wo a iables: Yes – emo ed in CRBSI, No – no emo ed in
CRABSI o no CRABSI. This is no a use ul p edic o clinically, because o he indi idual
pa ien he sou ce o he bac e emia is known and i unknown, he a iable is no ele an . To
app op ia ely analyse he associa ion be ween ca he e ex ac ion and mo ali y he analysis
needs o be limi ed o pa ien s wi h a CVC. I ecommend e ising model 2 o include a 3-
ca ego y a iable o CRBSI ex ac ed, CRBSI e ained and no CVC, o allow a mo e app op ia e
e alua ion o hese ac o s in a model including all indi idual componen s o complica ed
bac e emia. Al e na i ely, o concomi an ly, i would be o in e es o e alua e ca he e
ex ac ion in he subg oup o pa ien s wi h a CVC, among whom isk ac o s o mo ali y need
Page 4 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
o be e-e alua ed o selec he app op ia e con ounde s o he eg ession analysis including
ca he e ex ac ion.
Response: We ully ag ee; in ac , Model 2 was pe o med using pa ien s wi h CRBSI
wi hou ea ly ca he e emo al as he e e ence ca ego y o ea ly ca he e emo al
(es ima ion p o ided in he model) and o non-CRBSI (we did no include he es ima e
o non-CRBSI due o he lack o associa ion, bu we ag ee i should be included). We
now include his as Model 2a, and explained he s a i ica ion in he i le o Table 4. We
also add essed you second sugges ion abou adding an analysis o pa ien s wi h a CVC
and he e o e we added a Model 2b including only pa ien s wi h CRBSI.
Commen (6): Please de ine he a iables en e ed in o he eg ession analyses, hose emo ed in
he s epwise selec ion p ocess and all hose e ained wi h hei signi icance. Desc ibe he
a iables ha we e no en e ed due o co ela ions. Please documen he numbe o pa ien s
included in each eg ession model and he s a is ics o he model – i ness and calib a ion. See
ou guidance o epo ing o mul i a iable logis ic eg ession models.
h ps://www.clinicalmic obiologyandin ec ion.com/a icle/S1198-743X(19)30592-0/ ull ex . As
a e iewe no ed he model including all componen s o complica ed bac e emia migh ha e
highe dependen o independen a io han ecommended esul ing is o e i ing.
Response: Thank you o you aluable eedback. We added in he ex ha a de ailed
explana ion o models’ de elopmen is explained in he Supplemen a y ma e ial (please
see sec ion "Mul i a iable eg ession model me hodology").
Commen (7): Please include dispe sion measu es (SD, min-max, IQR, CI) o all con inuous
and ou comes and a io maeasu es epo ed in abs ac , ex and able and c ude numbe s
(nume a o / denomina o ) o all pe cen ages.
Response: We ha e included he dispe sion measu es (IQR and CI) whe e e hey we e
missing in he abs ac , ex , and ables. Addi ionally, we speci ied all absolu e numbe s
(nume a o /denomina o ) o each pe cen age.
Commen (8): App op ia e empi ic he apy was no associa ed wi h mo ali y in you coho .
You ollowing s a emen s in he discussion a e unsubs an ia ed: “Howe e , empi ic he apy
should be ini ia ed i isk ac o s o ad e se ou comes a e iden i ied” and lines 244-247
“Addi ionally, nea ly hal o he empi ic he apies we e ine ec i e agains he isola ed s ain,
unde sco ing he impo ance o s a ing ea men ha co e s me hicillin- esis an s ains when
CoNS in ec ions a e suspec ed. This app oach is suppo ed by ou inding ha o e 70% o
isola es in ou coho we e me hicillin- esis an , consis en wi h o he epo s.” Please e ise – I
Page 5 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
sugges dele ing hese pa s o he discussion. In addi ion, you no e o he i s ime in he
discussion ha only 90% p pa ien s ecei ed a ge ed he apy once cul u e esul s we e a ailable
– please include his a iable in he ables.
Response: We ag ee wi h he Edi o . We ha e made he eques ed changes by dele ing
he speci ied pa s o he discussion. Addi ionally, we ha e included he in o ma ion on
a ge ed he apy in he ables ha was missing.
Commen (9): Please check and e e ence lines 234-235 in he discussion on adhe ence o
guidelines ela ing o ca he e ex ac ion. Mos people ollow he IDSA guidelines ha as no ed
abo e allow e aining he ca he e in CONS CRBSI.
Response:
Following he Edi o ’s commen , we ha e comple ely e ised ha pa o he Discussion,
ecognizing ha IDSA guidelines conside ca he e e en ion in CoNS CRBSI.
Page 6 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
REVIEWER #1 COMMENTS
Commen (1): In his s udy, Spanish in es iga o s we e looking a pa ien s wi h coagulase-
nega i e s aphylococcus (CoNS) bac e emia in a p ospec i e s udy o bac e emia ha en olled a
mul iple hospi als in 2017. They sough o look a p edic o s o mo ali y and p oposed an
ope a ional de ini ion o complica ed bac e emia. Table 1 shows he cha ac e is ics o he 445
pa ien s included in he s udy. The au ho s de ined complica ed in ec ions acco ding o clinical
c i e ia p e iously applied o S. au eus bac e emia plus hey included lack o ca he e emo al
wi hin 48 hou s i CRBSI was p esen . 30 day mo ali y was highe in he complica ed
bac e emia g oup. Table 4 shows mul i- a ia e analysis whe e ea ly ca he e emo al was
associa ed wi h lowe mo ali y.
O e all: The au ho s a e add essing an impo an opic, namely bloods eam in ec ions due o
coagulase-nega i e s aphylococci. The main s eng hs o hei s udy a e he la ge sample size
and mul i-cen e design. The main weakness is ha i is obse a ional wi h all he d awbacks
such a s udy design en ails. Thus, al hough associa ions can be made, he clinical impac o his
s udy is limi ed. Mo eo e , he au ho s ha e combined in ec ion/pa ien cha ac e is ics, such as
pe sis en e e /p esence o in-dwelling de ice wi h p ac i ione decisions (i.e. ea ly ca he e
emo al) which ul ima ely obscu es hei de ini ion o “complica ed bac e emia”.
Response: Thank you o you commen . We acknowledge he signi ican limi a ions
inhe en in obse a ional s udies. Howe e , we belie e ha ou s udy design, ea u ing
p ospec i e da a collec ion, p o ides a aluable e lec ion o eal-wo ld p ac ices in
managing coagulase-nega i e s aphylococcal bloods eam in ec ions.
Rega ding he de ini ion o "complica ed bac e emia," we unde s and you conce n abou
he in eg a ion o pa ien /in ec ion cha ac e is ics and he physician's decision ega ding
ea ly ca he e emo al. We would like o emphasize ha ou de ini ion is no in ended o
be pa hophysiologic bu a he p ac ical o clinical use. I lack o ca he e emo al
impac s ou comes, we belie e ha conside ing i is use ul om a clinical pe spec i e.
Commen (2): The au ho s could ha e ac ually de i ed isk ac o s o poo ou come a he han
deciding on hem be o ehand and hen seeing i hey we e associa ed wi h mo ali y. This would
ha e s eng hened hei de ini ion o “complica ed in ec ion”.
Response: We app ecia e you insigh ul commen , which allows us o cla i y ou s udy's
pu pose. Ac ually, ou objec i e was hypo hesis-based, speci ically aiming o assess
whe he a p e-es ablished de ini ion o “complica ed bac e emia” was associa ed wi h an
inc eased isk o 30-day mo ali y ins ead o jus explo ing which a iables a e associa ed
wi h ou come. While iden i ying b oade isk ac o s o poo ou comes would indeed be
a ele an app oach, i aligns wi h a di e en objec i e han ou s. By applying p ede ined
c i e ia, we sough o e alua e hei associa ions wi h mo ali y, bo h collec i ely and
independen ly, wi hin ou ope a ional de ini ion.
Page 7 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
To e lec his mo e clea ly, we ha e e ised he s udy objec i es in he In oduc ion as
ollows: “This s udy aims o desc ibe a la ge coho o CoNS BSI cases, es whe he
p ede ined c i e ia o complica ed bac e emia a e associa ed wi h 30-day mo ali y and
iden i y o he ou come p edic o s.”
Commen (3): The e a e s a is ical e o s, such as in Table 1 when he au ho s compa e
uncomplica ed and complica ed in ec ion in e ms o whe he pa ien s ha e ea ly ca he e
emo al. Gi en ha ea ly ca he e emo al is pa o he de ini ion, i is no su p ising ha he e
is a s a is ically signi ican di e ence be ween he wo g oups. Gi en ha his manusc ip has
complex s a is ics a i s hea , his kind o compa ison aises conce n o he alidi y o he o he
indings.
Response: The Re iewe is igh ; we dele ed he s a is ical compa isons in Table 1
dealing wi h ac o s which a e c i e ia o complica ed bac e emia, as pa ien s wi h
uncomplica ed bac e emia canno mee hese c i e ia. We also ook his oppo uni y o
echeck all he da a be o e esubmission. We also cla i y ha in he Table oo .
Commen (4): On line 233 he au ho s s a e ha lack o ea ly emo al o a ca he e in CoNS
CRBSI indica es poo adhe ence o CRBSI managemen s anda ds wi hou gi ing a e e ence –
howe e , he 2009 IDSA guidelines (which a e mos cu en ) s a e ha is app op ia e o lea e a
ca he e in place ini ially o CoNS CRBSI and see how pa ien s do. The as majo i y o
pa ien s a e cu ed wi h his app oach (see PMID 19780661 which is no e e enced he e) which
in u n sa es pa ien s om unde going unnecessa y ca he e change. Thus, I would sugges he
au ho s empe his c i icism unless hey ha e o he da a o suppo i .
Response: Thank you o you insigh ul commen ; we ully ag ee wi h you
obse a ions. This conce n was also aised by he Edi o . We app ecia e you sugges ion
ega ding he e e ence, which has helped us cla i y ou pe spec i e. Following you and
he Edi o ’s commen , we ha e e ised in deep he Discussion sec ion dealing wi h his
aspec .
Commen (5): The au ho s should be ca e ul abou including ca he e e en ion in any model
wi hou speci ically unde s anding why ca he e s we e e ained. Fo example, i a pa ien is
dying o ano he p ocess, ca he e e en ion may simply e lec he unwillingness o he ea ing
physicians o pu he pa ien h ough mo e p ocedu es.
Response: Thank you o you commen . We ag ee ha unde s anding he speci ic easons o
ca he e e en ion is impo an ; howe e , ou da ase does no include his in o ma ion. Despi e
his limi a ion, he associa ion o ea ly ca he e emo al wi h mo ali y in ou coho emains
Page 8 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
Commen (6): Majo poin s. Me hods. Lines 142-144: How many a iables wen in o he
mul i a ia e Cox eg ession analysis? Was he numbe in p opo ion o 77 e en s/dea hs a
day30? Please cla i y and commen .
Response: Thank you o aising his poin , which was also highligh ed in Commen (6)
om he Edi o . We hope ha ou modi ica ions ha e cla i ied his aspec .
Commen (7): Majo poin . Line 144: In e ac ions we e e alua ed – could he au ho s please
speci y his impo an s ep in he espec i e s eps.
Response:
In e ac ions be ween complica ed bac e emia and he o he a iables in he inal model
we e assessed, as well as in e ac ions be ween he a iables and each indi idual c i e ion
de ining complica ed bac e emia. No clinically o s a is ically signi ican in e ac ions
we e obse ed. The explana ion can now be ead in he Sec ion “Me hods [3].
Mul i a iable eg ession model me hodology” in Supplemen a y ma e ial.
Commen (8): Majo poin . Resul s. I ‘me as a ic complica ion’ occu ed in <3% o cases,
could his c i e ion be omi ed?
Response: Thank you o he commen . Since he de ini ion o complica ed bac e emia
was p e-de ined, we decided o e ain all he c i e ia conside ed. To no e, 5 o he 7
pa ien s wi h me as a ic complica ions me o he c i e ia o "complica ed bac e emia": 1
was a ascula p os he ic de ice ca ie , 1 had an a icula de ice, 2 had pe sis en e e ,
and 1 had a CRBSI wi hou ea ly ca he e emo al. Only wo pa ien s did no mee any
o he "complica ed c i e ia," and he de ini ion was solely a ibu ed o he p esence o
me as a ic oci. We ha e added his explana ion in he Supplemen a y Ma e ial unde he
sec ion "Me hods [2]: Jus i ica ion o he inclusion o “me as a ic complica ions” as a
c i e ion o “complica ed bac e emia”."
Commen (9): Mino poin . Gene al aspec s. The manusc ip could bene i om some language
edi ing.
Response: Thank you o you eedback. We ha e e ised he manusc ip o cla i y and
language quali y o enhance i s o e all eadabili y.
Commen (10): Mino poin . Abs ac and h oughou he manusc ip . Sugges o use ‘in ec i e’
endoca di is ins ead o ‘in ec ious’.
Page 15 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
Response: Thank you o he sugges ion. We ag ee and ha e made he modi ica ion o
use "in ec i e" endoca di is ins ead o "in ec ious" h oughou he manusc ip and
addi ional ma e ial.
Commen (11): Mino poin . Line 103: which (in a ascula ) ca he e s we e included – please
speci y.
Response: Cul u es d awn om any ype o in a ascula ca he e could be sen o
mic obiology o iden i ica ion o CRBSI, i submi ed alongside pe iphe al blood
cul u es. This in o ma ion was added o he Supplemen a y Ma e ial unde he sec ion
“S udy mic obiological and clinical inclusion c i e ia o CoNS bloods eam in ec ion
episodes,” speci ically wi hin he mic obiological c i e ia.
Commen (12): Mino poin . Me hods and Resul s. Please speci y which implan ed de ices we e
included (so a ou lined in able, bu no in ex ).
Response: Since his in o ma ion is in he Table, we hink is i su icien , also gi en he
wo ds limi ; howe e , we a e open o ollowing any speci ic guidance om he Edi o on
his ma e .
Commen (13): Mino poin . Discussion. Line 214: please elabo a e o gi e e e ences o he
‘p e ious s udies’ men ioned.
Response: The e e ence was indeed p o ided a he end o he sen ence, bu we will
mo e i wi hin he sen ence o imp o ed cla i y.
We would inally like o decla e he ollowing:
•In Table 2, he c i e ion o “p ima y endoca di is” was omi ed in he p e ious e sion.
While i was conside ed in he coun , i was inad e en ly excluded om he able. This
has been co ec ed, and he c i e ion is now included in he e ised able.
•In esponse o he au ho s' and e iewe s' commen s, he cu en e sion o he
manusc ip is 2,593 wo ds, which sligh ly exceeds he wo d limi .
Page 16 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
1
1Mo ali y p edic o s and de ini ion p oposal o complica ed coagulase-nega i e
S aphylococcus
2bac e emia. A mul icen e p ospec i e coho s udy
3
4Au ho s
5Benede a Va isco1, Ped o Ma ía Ma ínez Pé ez-C espo2, Pila Re ama -Gen il3,4, Inmaculada López
6He nandez3,4, Mª Ca men Fa iñas-Ál a ez4,5, Isabel Fe nández-Na al6, Ma ía Te esa Pé ez-
7Rod íguez7, Ane Josune Goikoe xea Agui e8, Juan Manuel Sánchez-Cal o9, Luis Buzón Ma ín10,
8E a León-Jiménez2, Da id Vinuesa Ga cía11, José Ma ía Regue a-Iglesias12, Albe o Bahamonde-
9Ca asco13, Jona han Fe nández Suá ez14, Jesús Rod íguez-Baño3,4*, Luis Edua do López-Co és3,4*,
10 on behal o he PROBAC REIPI/GEIH-SEIMC/SAEI G oup
11
12 A ilia ions
13 1Depa men o Heal h Sciences, Clinic o In ec ious Diseases, Uni e si y o Milan, ASST San i Paolo
14 e Ca lo, Via A. Di Rudinì, 8, 20142, Milan, I aly.
15 2Unidad de En e medades In ecciosas y Mic obiología, Hospi al Uni e si a io de Valme, 41014,
16 Se illa, Spain.
17 3Unidad Clínica de En e medades In ecciosas y Mic obiología, Hospi al Uni e si a io Vi gen
18 Maca ena; Depa amen os de Medicina y Mic obiología, Facul ad de Medicina, Uni e sidad de
19 Se illa; Ins i u o de Biomedicina de Se illa (IBiS)/CSIC, 41009, Se ille, Spain.
20 4CIBERINFEC, Ins i u o de Salud Ca los III, 28029, Mad id, Spain.
21 5Unidad de En e medades In ecciosas, Hospi al Uni e si a io Ma qués de Valdecilla, Uni e sidad de
22 Can ab ia, IDIVAL, 39008, San ande , Spain.
Page 17 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
2
23 6Depa men o Clinical Mic obiology, Complejo Asis encial Uni e si a io de León (CAULE), 24071,
24 León, Spain.
25 7 In ec ious Diseases Uni , Depa men o In e nal Medicine, Complexo Hospi ala io Uni e si a io de
26 Vigo, 36312, Vigo, Spain; Ins i u o de In es igación Biomédica Galicia Su , Spain.
27 8 Unidad de En e medades In ecciosas, Hospi al Uni e si a io de C uces, 48903, Bizkaia, Spain.
28 9 Uni o In ec ious Diseases and Clinical Mic obiology. Je ez De La F on e a Uni e si y Hospi al,
29 11407, Je ez De La F on e a, Cádiz, Spain.
30 10 Unidad de Ges ión Clínica de En e medades In ecciosas, Hospi al Uni e si a io de Bu gos, 09006,
31 Bu gos, Spain.
32 11 Unidad Ges ión Clínica En e medades In ecciosas, Hospi al Uni e si a io Clínico San Cecilio,
33 18016, G anada, Spain.
34 12Se icio de En e medades In ecciosas, Hospi al Regional Uni e si a io de Málaga, IBIMA Málaga,
35 29010, Málaga, Spain.
36 13Depa amen o de Medicina In e na, Hospi al de El Bie zo, 24404, Pon e ada, Spain.
37 14Unidad de Mic obiología, Ins i u o de In es igación Sani a ia del P incipado de As u ias (ISPA),
38 Hospi al Uni e si a io Cen al de As u ias, 33011, O iedo, Spain.
39 *These au ho s con ibu ed equally as senio au ho s.
40
41 Co esponding au ho
42 Luis Edua do López-Co és
43 In ec ious Diseases and Mic obiology Uni
44 Hospi al Uni e si a io Vi gen Maca ena
45 Depa men o Medicine, Uni e si y o Se illa
Page 18 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
3
46 Biomedicines Ins i u e o Se illa
47 41009 Se illa, Spain
48 Phone: +34 670 946 430
49 Email: [email p o ec ed]
Page 19 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
4
50 Keywo ds. Coagulase-nega i e s aphylococci; bloods eam in ec ions; complica ed bac e emia
51 A icle ype. O iginal a icle
52 Subjec sec ion. Bac e ial In ec ions; An imic obial S ewa dship
53 Abs ac wo d coun . 245
54 Tex wo d coun . 2593
55
Page 20 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
5
56 ABSTRACT
57 Objec i e. To explo e a de ini ion o complica ed coagulase-nega i e s aphylococci bloods eam
58 in ec ions (CoNS BSI), and o iden i y p edic o s o mo ali y.
59
60 Me hods. P ospec i e coho s udy conduc ed om Oc obe 2016 o Ma ch 2017 in 26 Spanish
61 hospi als. Complica ed CoNS BSI c i e ia included lack o ea ly ca he e emo al in ca he e - ela ed
62 cases, o eign indwelling implan , pe sis en bac e emia, e e ≥72 hou s on ac i e he apy, me as a ic
63 in ec ion o deep-sea ed ocus and in ec i e endoca di is. Independen p edic o s o 30-day mo ali y
64 we e e alua ed by Cox eg ession, and he impac o he de ini ion o complica ed bac e emia assessed.
65
66 Resul s. O e all, 445 CoNS BSI cases we e included; ca he e - ela ed in ec ions we e p edominan
67 (336/445, 75.5%). Complica ed bac e emia was iden i ied in 240/445 pa ien s (53.9%); 30-day
68 mo ali y in complica ed and uncomplica ed cases we e 53/240 (22.1%) and 24/205 (11.7%),
69 espec i ely (p=0.004). P edic o s o 30-day mo ali y iden i ied in he mul i a ia e analysis included
70 age (HR 1.03, 95%CI 1.01-1.05), ce eb o ascula disease (HR 2.58, 95%CI 1.45-4.58),
71 immunosupp essi e he apy (HR 2.16, 95%CI 1.22-3.84), SOFA sco e (HR 1.09, 95%CI 1.03-1.16),
72 and complica ed bac e emia (HR 2.14, 95%CI 1.29-3.53). A ca he e - ela ed sou ce o bac e emia was
73 ound o be p o ec i e (HR 0.49, 95%CI 0.30-0.80). When speci ic c i e ia o de ine complica ed
74 bac e emia we e included, e e ≥72h was associa ed wi h inc eased isk o dea h (HR 2.52, 95%CI
75 1.52-4.17) and ea ly ca he e emo al was p o ec i e (HR 0.47, 95%CI 0.26-0.83).
76
Page 21 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
6
77 Conclusions. A high p opo ion o pa ien s p esen ed complica ed bac e emia acco ding o he
78 p oposed c i e ia; hese pa ien s had highe haza ds o mo ali y. O he mo ali y p edic o s we e
79 iden i ied. Fu he s udies would be needed o alida e he p oposed c i e ia.
Page 22 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
7
80 INTRODUCTION
81
82 Coagulase-nega i e s aphylococci (CoNS) ha e adi ionally been ega ded as low- i ulence
83 o ganisms due o hei ole as skin commensals and equen con aminan s in mic obiological
84 samples. They a e now ecognized as a leading cause o nosocomial bloods eam in ec ions (BSI),
85 pa icula ly ca he e - ela ed (CRBSI), due o hei abili y o o m bio ilms.1 Despi e hei high
86 p e alence, esea ch in o he mo bidi y and mo ali y o CoNS bac e emia emains limi ed, wi h
87 epo ed mo ali y a es anging om 4% o 24%.2–7 Fu he mo e, ea men ecommenda ions ely
88 on limi ed e idence, e lec ing he challenge o dis inguishing ue in ec ions om con amina ion.8
89 Ne e heless, a combina ion o accu a e clinical e alua ion and in e p e a ion o mic obiological
90 indings (pa icula ly he numbe o posi i e blood cul u es)9 can help educe misdiagnoses.
91 In e ms o ea men , he In ec ious Diseases Socie y o Ame ica (IDSA) guidelines10
92 ecommend ea ing uncomplica ed CoNS CRBSI o 5-7 days i he ca he e is emo ed, and o 10-
93 14 days wi h an ibio ic lock he apy i he ca he e is e ained (le el BIII e idence). Howe e , o he
94 bes o ou knowledge, he impac o delayed ca he e emo al in high- isk pa ien s has no been
95 sys ema ically assessed. Addi ionally, c i e ia o complica ed bac e emia used in he guidelines
96 include suppu a i e h ombophlebi is, endoca di is, os eomyeli is, and me as a ic seeding, hough
97 hese c i e ia a e no speci ic o CoNS in ec ions.10 Whe he addi ional c i e ia, such as p olonged
98 e e , pe sis en bac e emia, and lack o sou ce con ol (conside ed o
S aphylococcus au eus
)11,12
99 a e associa ed wi h wo se ou comes in CoNS BSI emains uns udied.
100 This s udy aims o desc ibe a la ge coho o CoNS BSI cases, e alua e he associa ion
101 be ween p ede ined c i e ia o complica ed bac e emia and 30-day mo ali y, and iden i y o he
102 ou come p edic o s.
Page 23 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
8
103
104 METHODS
105
106
Design and se ing
107 This s udy is pa o he PROBAC p ojec , a p ospec i e coho s udy in ol ing pa ien s >14
108 yea s admi ed o 26 hospi als in Spain om Oc obe 1, 2016, o Ma ch 31, 2017, wi h clinically
109 signi ican BSI.13,14 The coho included 18 e ia y ca e and 8 communi y hospi als.
110
111
Inclusion and exclusion c i e ia
112 To ensu e he clinical signi icance o CoNS bloods eam in ec ions (BSI), episodes we e
113 included in he s udy i hey me bo h mic obiological and clinical p especi ied c i e ia (Me hods [1],
114 Supplemen a y Ma e ial). In summa y, mic obiological c i e ia equi ed he isola ion o he same
115 CoNS species wi h an iden ical suscep ibili y p o ile om a leas wo dis inc blood cul u e se s, wi h
116 addi ional c i e ia o ca he e - ela ed in ec ions.10 Clinical c i e ia included he p esence o a leas
117 wo sys emic in ec ion signs ( e e o hypo he mia, achyca dia, achypnoea, leukocy osis o
118 leukopenia). Addi ionally, o he causes o in ec ion we e easonably excluded; o pa ien s in whom
119 he sou ce o bac e emia was excep ional o CoNS (i.e., pneumonia, in a-abdominal in ec ion,
120 meningi is, o unknown sou ce in pa ien s wi hou ca he e s o p os he ic ma e ial), he opinion o a
121 second local in es iga o was sough .
122 The ollowing exclusion c i e ia we e applied: polymic obial BSI; in ec ions by
123
S aphylococcus lugdunensis
; 15 ecu en episodes o bac e emia, de ined by he isola ion o he same
124 species wi h clinical signs and symp oms consis en wi h in ec ion, wi hin h ee mon hs a e esolu ion
125 o he ini ial episode.
Page 24 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
15
261 disease, and immunosupp essi e he apy we e linked o highe mo ali y, likely e lec ing pa ien
262 ail y, while he SOFA sco e indica ed acu e se e i y. Addi ionally, ca he e - ela ed in ec ions we e
263 associa ed wi h lowe mo ali y compa ed o in ec ions om o he sou ces.
264 In he absence o speci ic c i e ia o CoNS BSI, ou s udy used modi ied c i e ia o iginally
265 designed o
S. au eus
complica ed bac e emia. The inding ha complica ed bac e emia is
266 independen ly associa ed wi h highe mo ali y sugges s ha hese c i e ia migh be use ul o CoNS
267 BSI as well. Howe e , he sepa a e analysis in he second model, whe e no all c i e ia we e linked o
268 mo ali y, indica es he need o a ailo ed app oach o CoNS BSI in u u e esea ch.
269 FUBC we e pe o med in ewe han 30% o cases in ou coho . Al hough speci ic guidelines
270 a e lacking, ob aining hem in high- isk pa ien s wi h complica ed bac e emia may be bene icial,
271 especially gi en he obse ed associa ion be ween mo ali y and pe sis en e e , which could indica e
272 ongoing bac e emia. In con as , FUBC may be unnecessa y o low- isk cases, as sugges ed by Badia-
273 Cebada e al.23 Thei ial compa ing an ibio ics e sus no an ibio ics in low- isk CoNS CRBSI ound
274 only one case o pe sis en bac e emia in he no-an ibio ics g oup. Howe e , he small sample size due
275 o s udy in e up ion limi s he s eng h o hese indings.
276 Al hough ou s udy ound ea ly ca he e emo al o be p o ec i e agains mo ali y, i was
277 ca ied ou in ewe han 60% o CRBSI cases. IDSA guidelines conside he possibili y o ca he e
278 e en ion in uncomplica ed CoNS CRBSI cases.10 Howe e , despi e he ac ha some s udies sugges
279 ha ca he e e en ion in CoNS bloods eam in ec ions may no signi ican ly impac mo ali y, i
280 emains an independen isk ac o o ecu ence,24 while ecu ence is a e a e ca he e emo al.25
281 Ou inding ha ea ly ca he e emo al is independen ly associa ed wi h a p o ec i e e ec on 30-day
282 mo ali y sugges s he need o u he s udies o be e iden i y he subg oups o pa ien s in whom
283 e aining he ca he e is sa e.
Page 31 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
16
284 In ou s udy, abou one- hi d o pa ien s did no ecei e empi ic he apy a he onse o in ec ion;
285 howe e , app oxima ely 90% o hese pa ien s we e ini ia ed on a ge ed he apy once cul u e esul s
286 became a ailable. Omi ing empi ic he apy may be jus i ied i no isk ac o s a e p esen and sou ce
287 con ol (e.g. ca he e emo al) is achie ed, po en ially making subsequen a ge ed he apy
288 unnecessa y. 23,26
289 Ou s udy has limi a ions. Fi s , he absence o a compa a i e g oup es ic s ou abili y o
290 es ablish causal ela ionships and assess he p opo ion o mo ali y di ec ly a ibu able o CoNS BSI
291 e sus unde lying condi ions. Al hough a mino i y o pa ien s did no comple e he 30-day ollow-up,
292 sensi i i y analyses sugges he esul s a e obus . We acknowledge ha posi i e FUBC esul s likely
293 ep esen only a subse o pe sis en bac e emia cases, as some pa ien s wi hou FUBC may ha e had
294 unde ec ed posi i e esul s; al hough we e ospec i ely analyzed ac o s in luencing he decision o
295 pe o m FUBC, he da a lacks clea indica ions o each case. None heless, we belie e including his
296 a iable is impo an o highligh he need o u he s udies o cla i y FUBC indica ions in CoNS
297 bloods eam in ec ions. Managemen o pa ien s was pe o med acco ding o a ending physicians and
298 he e o e was no s anda dized; none heless, we belie e he da a e lec eal-wo ld clinical p ac ice.
299 Despi e e o s like in es iga o aining and da a moni o ing, some da a collec ion e o s may s ill
300 exis . Howe e , he s udy's s eng hs include a la ge sample size, and a mul icen e , p ospec i e design
301 conduc ed by in ec ious disease expe s in hospi als wi h bac e emia p og ams, which enhances he
302 c edibili y o ou indings.
303 In summa y, ou s udy iden i ies key mo ali y ac o s in CoNS BSI and emphasizes he need
304 o imp o ed isk s a i ica ion. The p oposed de ini ion o complica ed CoNS BSI needs alida ion
305 in u he s udies.
306
Page 32 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
17
307 AUTHOR CONTRIBUTIONS
308
309 Concep ualiza ion: LELC, JRB. Me hodology and s a is ical analysis: BV, LELC, JRB.
310 O iginal d a : BV. Re iew & edi ing: BV, LELC, JRB (all au ho s). Supe ision: JRB. P ojec
311 adminis a ion: LELC. Funding acquisi ion: LELC, JRB. All au ho s con ibu ed o he inal
312 manusc ip .
313
314 FUNDING
315
316 This esea ch was suppo ed by g an s om Plan Nacional de I+D+i 2013-2016, Ins i u o de
317 Salud Ca los III, Subdi ección Gene al de Redes y Cen os de In es igación Coope a i a, Minis e io
318 de Ciencia, Inno ación y Uni e sidades (PI16/01432); he Spanish Ne wo k o Resea ch in In ec ious
319 Diseases (REIPI) (RD16/0016/0001; RD16/ 0016/0008); and co- inanced by he Eu opean Regional
320 De elopmen Fund unde he 'A way o achie e Eu ope' Ope a ional P og am o In elligen G ow h
321 2014-2020.
322
323 TRANSPARENCY STATEMENT
324
325 PRG has se ed as a scien i ic ad iso o Shionogi and Ad anz, speake o Angelini, and
326 Mena ini. LELC has se ed as a scien i ic ad iso o No a is, speake o MSD, P ize , Angelini,
327 and ViiV, and aine o MSD and ViiV. The o he au ho s decla e no con lic s o in e es .
Page 33 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
18
328 REFERENCES
329 1. Heilmann C, Ziebuh W, Becke K. A e coagulase-nega i e s aphylococci i ulen ? Clinical
330 Mic obiology and In ec ion. 2019;25(9):1071-1080. doi:10.1016/j.cmi.2018.11.012
331
332 2. Molina J, Peñuela I, Lepe JA, e al. Mo ali y and hospi al s ay ela ed o coagulase-nega i e
333 S aphylococci bac e emia in non-c i ical pa ien s. Jou nal o In ec ion. 2013;66(2):155-162.
334 doi:10.1016/j.jin .2012.10.021
335
336
337 3. Pa k SY, Kwon KH, Chung JW, Huh HJ, Chae SL. Coagulase-nega i e s aphylococcal
338 bac e emia: isk ac o s o mo ali y and impac o ini ial app op ia e an imic obial he apy
339 on ou come. Eu opean Jou nal o Clinical Mic obiology and In ec ious Diseases.
340 2015;34(7):1395-1401. doi:10.1007/s10096-015-2364-3
341
342 4. Hen zien M, S ady C, Ve ne -Ga nie V, e al. P ognos ic ac o s associa ed wi h 30-day in-
343 hospi al mo ali y in coagulase-nega i e S aphylococcus bac e aemia: no impac o
344 ancomycin minimum inhibi o y concen a ion. In ec Dis. 2017;49(9):664-673.
345 doi:10.1080/23744235.2017.1323346
346
347 5. Cui J, Liang Z, Mo Z, Zhang J. The species dis ibu ion, an imic obial esis ance and isk
348 ac o s o poo ou come o coagulase-nega i e s aphylococci bac e aemia in China.
349 An imic ob Resis In ec Con ol. 2019;8(1):1-10. doi:10.1186/s13756-019-0523-5
350
351 6. Yamada K, Namikawa H, Fujimo o H, e al. Clinical cha ac e is ics o me hicillin- esis an
352 coagulase-nega i e s aphylococcal bac e emia in a e ia y hospi al. In e nal Medicine.
353 2017;56(7):781-785. doi:10.2169/in e nalmedicine.56.7715
354
355 7. Rosa RG, Dos San os RP, Goldani LZ. Mo ali y ela ed o coagulase-nega i e s aphylococcal
356 bac e emia in eb ile neu openia: A coho s udy. Canadian Jou nal o In ec ious Diseases
357 and Medical Mic obiology. 2014;25(1):14-18. doi:10.1155/2014/702621
358
359 8. Rahkonen M, Lu inen S, Koskela M, Hau ala T. T ue bac e emias caused by coagulase
360 nega i e S aphylococcus a e di icul o dis inguish om blood cul u e con aminan s.
361 Eu opean Jou nal o Clinical Mic obiology and In ec ious Diseases. 2012;31(10):2639-2644.
362 doi:10.1007/s10096-012-1607-9
363
364 9. Mi e S, Weins ein MP, Reime LG, Wilson ML, Relle LB. Rele ance o he numbe o
365 posi i e bo les in de e mining clinical signi icance o coagulase-nega i e s aphylococci in
366 blood cul u es. J Clin Mic obiol. 2001;39(9):3279-3281. doi:10.1128/JCM.39.9.3279-
367 3281.2001
368
369 10. Me mel LA, Allon M, Bouza E, e al. Clinical p ac ice guidelines o he diagnosis and
370 managemen o in a ascula ca he e - ela ed in ec ion: 2009 upda e by he In ec ious
371 Diseases Socie y o Ame ica. Clinical In ec ious Diseases. 2009;49(1):1-45.
372 doi:10.1086/599376
373
374 11. Jung N, Rieg S. Essen ials in he managemen o S. au eus bloods eam in ec ion. In ec ion.
375 2018;46(4):441-442. doi:10.1007/s15010-018-1130-8
376
Page 34 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
19
377 12. López-Co és LE, Del To o MD, Gál ez-Acebal J, e al. Impac o an e idence-based bundle
378 in e en ion in he quali y-o -ca e managemen and ou come o S aphylococcus au eus
379 bac e emia. Clinical In ec ious Diseases. 2013;57(9):1225-1233. doi:10.1093/cid/ci 499
380
381 13. Calo F, Re ama P, Pe ez-C espo PMM, e al. Ca he e - ela ed bloods eam in ec ions:
382 P edic i e ac o s o G am-nega i e bac e ia ae iology and 30 day mo ali y in a mul icen e
383 p ospec i e coho . Jou nal o An imic obial Chemo he apy. 2020;75(10):3056-3061.
384 doi:10.1093/jac/dkaa262
385
386 14. Mussa M, Ma ínez Pé ez-C espo PM, Lopez-Co es LE, e al. Risk Fac o s and P edic i e
387 Sco e o Bac e emic Bilia y T ac In ec ions Due o En e ococcus aecalis and En e ococcus
388 aecium: a Mul icen e Coho S udy om he PROBAC P ojec . Mic obiol Spec .
389 2022;10(4). doi:10.1128/spec um.00051-22
390
391 15. F ank KL, Del Pozo JL, Pa el R. F om clinical mic obiology o in ec ion pa hogenesis: How
392 da ing o be di e en wo ks o S aphylococcus lugdunensis. Clin Mic obiol Re .
393 2008;21(1):111-133. doi:10.1128/CMR.00036-07
394
395 16. Rod íguez Díaz JC, Guna Se ano M del R, La osa Esca ín N, Ma ín A iaza M. Diagnós ico
396 mic obiológico de la bac e iemia y la ungemia: hemocul i os y mé odos molecula es.
397 Sociedad Española de En e medades In ecciosas y Mic obiología Clínica. 2017. Accessed
398 No embe 6, 2024.
399 h ps://seimc.o g/con enidos/documen oscien i icos/p ocedimien osmic obiologia/seimc-
400 p ocedimien omic obiologia62.pd
401
402 17. Cha lson ME, Pompei P, Ales KL, MacKenzie CR. A new me hod o classi ying p ognos ic
403 como bidi y in longi udinal s udies: De elopmen and alida ion. J Ch onic Dis.
404 1987;40(5):373-383. doi:h ps://doi.o g/10.1016/0021-9681(87)90171-8.
405
406 18. F iedman ND, Kaye KS, S ou JE, e al. Heal h ca e-associa ed bloods eam in ec ions in
407 adul s: A eason o change he accep ed de ini ion o communi y-acqui ed in ec ions. Ann
408 In e n Med. 2002;137(10):791-797. doi:10.7326/0003-4819-137-10-200211190-00007
409
410 19. Hil M, Yu VL, Sha p J, Zu a le JJ, Ko ick JA, Mude RR. An ibio ic he apy o
411 Pseudomonas ae uginosa bac e emia: Ou come co ela ions in a p ospec i e s udy o 200
412 pa ien s. Am J Med. 1989;87(5):540-546. doi:10.1016/S0002-9343(89)80611-4
413
414 20. Fe nando SM, Rochwe g B, Seely AJE. Clinical implica ions o he hi d in e na ional
415 consensus de ini ions o sepsis and sep ic shock (Sepsis-3). JAMA. 2018;190(36):E1058-
416 E1059. doi:10.1503/cmaj.170149
417
418 21. Cha es F, Ga nacho-Mon e o J, del Pozo JL, e al. Diagnosis and ea men o ca he e - ela ed
419 bloods eam in ec ion: Clinical guidelines o he Spanish Socie y o In ec ious Diseases and
420 Clinical Mic obiology and (SEIMC) and he Spanish Socie y o Spanish Socie y o In ensi e
421 and C i ical Ca e Medicine a. Med In ensi a. 2018;42(1):5-36.
422 doi:10.1016/j.medin.2017.09.012
423
424 22. O ihuela-Ma ín J, Rod íguez-Núñez O, Mo a a L, e al. Pe o mance o di e en ial ime o
425 posi i i y as a ou ine diagnos ic es o ca he e - ela ed bloods eam in ec ions: a single-
426 cen e expe ience. Clinical Mic obiology and In ec ion. 2020;26(3):383.e1-383.e7.
Page 35 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
20
427 doi:10.1016/j.cmi.2019.07.001
428
429 23. Badia-Cebada L, Ca mezim J, Pé ez-Rod íguez MT, e al. Randomized Clinical T ial o he
430 Need o An ibio ic T ea men o Low-Risk Ca he e -Rela ed Bloods eam In ec ion Caused
431 by Coagulase-Nega i e S aphylococci. An ibio ics. 2023;12(5).
432 doi:10.3390/an ibio ics12050839
433
434 24. Raad I, Kassa R, Ghannam D, Cha a i AM, Hachem R, Jiang Y. Managemen o he ca he e
435 in documen ed ca he e - ela ed coagulase-nega i e s aphylococcal bac e emia: Remo e o
436 e ain? Clinical In ec ious Diseases. 2009;49(8):1187-1194. doi:10.1086/605694
437
438 25. Hebeisen U, Babouee Flu y B, A kinson A, Ma schall J, Bue i N. Ca he e - ela ed
439 bloods eam in ec ions due o coagulase-nega i e s aphylococci managed wi h ca he e
440 emo al: Recu ences a e a e. Am J In ec Con ol. 2020;48(7):837-839.
441 doi:10.1016/j.ajic.2019.10.013
442
443 26. Hebeisen UP, A kinson A, Ma schall J, Bue i N. Ca he e - ela ed bloods eam in ec ions wi h
444 coagulase-nega i e s aphylococci: A e an ibio ics necessa y i he ca he e is emo ed?
445 An imic ob Resis In ec Con ol. 2019;8(1):1-8. doi:10.1186/s13756-019-0474-x
446
Page 36 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
1
1Mo ali y p edic o s and de ini ion p oposal o complica ed coagulase-nega i e
S aphylococcus
2bac e emia. A mul icen e p ospec i e coho s udy
3
4Au ho s
5Benede a Va isco1, Ped o Ma ía Ma ínez Pé ez-C espo2, Pila Re ama -Gen il3,4, Inmaculada López
6He nandez3,4, Mª Ca men Fa iñas-Ál a ez4,5, Isabel Fe nández-Na al6, Ma ía Te esa Pé ez-
7Rod íguez7, Ane Josune Goikoe xea Agui e8, Juan Manuel Sánchez-Cal o9, Luis Buzón Ma ín10,
8E a León-Jiménez2, Da id Vinuesa Ga cía11, José Ma ía Regue a-Iglesias12, Albe o Bahamonde-
9Ca asco13, Jona han Fe nández Suá ez14, Jesús Rod íguez-Baño3,4*, Luis Edua do López-Co és3,4*,
10 on behal o he PROBAC REIPI/GEIH-SEIMC/SAEI G oup
11
12 A ilia ions
13 1Depa men o Heal h Sciences, Clinic o In ec ious Diseases, Uni e si y o Milan, ASST San i Paolo
14 e Ca lo, Via A. Di Rudinì, 8, 20142, Milan, I aly.
15 2Unidad de En e medades In ecciosas y Mic obiología, Hospi al Uni e si a io de Valme, 41014,
16 Se illa, Spain.
17 3Unidad Clínica de En e medades In ecciosas y Mic obiología, Hospi al Uni e si a io Vi gen
18 Maca ena; Depa amen os de Medicina y Mic obiología, Facul ad de Medicina, Uni e sidad de
19 Se illa; Ins i u o de Biomedicina de Se illa (IBiS)/CSIC, 41009, Se ille, Spain.
20 4CIBERINFEC, Ins i u o de Salud Ca los III, 28029, Mad id, Spain.
21 5Unidad de En e medades In ecciosas, Hospi al Uni e si a io Ma qués de Valdecilla, Uni e sidad de
22 Can ab ia, IDIVAL, 39008, San ande , Spain.
Page 37 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
2
23 6Depa men o Clinical Mic obiology, Complejo Asis encial Uni e si a io de León (CAULE), 24071,
24 León, Spain.
25 7 In ec ious Diseases Uni , Depa men o In e nal Medicine, Complexo Hospi ala io Uni e si a io de
26 Vigo, 36312, Vigo, Spain; Ins i u o de In es igación Biomédica Galicia Su , Spain.
27 8 Unidad de En e medades In ecciosas, Hospi al Uni e si a io de C uces, 48903, Bizkaia, Spain.
28 9 Uni o In ec ious Diseases and Clinical Mic obiology. Je ez De La F on e a Uni e si y Hospi al,
29 11407, Je ez De La F on e a, Cádiz, Spain.
30 10 Unidad de Ges ión Clínica de En e medades In ecciosas, Hospi al Uni e si a io de Bu gos, 09006,
31 Bu gos, Spain.
32 11 Unidad Ges ión Clínica En e medades In ecciosas, Hospi al Uni e si a io Clínico San Cecilio,
33 18016, G anada, Spain.
34 12Se icio de En e medades In ecciosas, Hospi al Regional Uni e si a io de Málaga, IBIMA Málaga,
35 29010, Málaga, Spain.
36 13Depa amen o de Medicina In e na, Hospi al de El Bie zo, 24404, Pon e ada, Spain.
37 14Unidad de Mic obiología, Ins i u o de In es igación Sani a ia del P incipado de As u ias (ISPA),
38 Hospi al Uni e si a io Cen al de As u ias, 33011, O iedo, Spain.
39 *These au ho s con ibu ed equally as senio au ho s.
40
41 Co esponding au ho
42 Luis Edua do López-Co és
43 In ec ious Diseases and Mic obiology Uni
44 Hospi al Uni e si a io Vi gen Maca ena
45 Depa men o Medicine, Uni e si y o Se illa
Page 38 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
3
46 Biomedicines Ins i u e o Se illa
47 41009 Se illa, Spain
48 Phone: +34 670 946 430
49 Email: [email p o ec ed]
Page 39 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
4
50 Keywo ds. Coagulase-nega i e s aphylococci; bloods eam in ec ions; complica ed bac e emia
51 A icle ype. O iginal a icle
52 Subjec sec ion. Bac e ial In ec ions; An imic obial S ewa dship
53 Abs ac wo d coun . 245
54 Tex wo d coun . 2593
55
Page 40 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
11
171 implan , pe sis en bac e emia, e e ≥72 hou s on ea men , me as a ic in ec ion o deep-sea ed ocus,
172 in ec ious in ec i e endoca di is ( hese we e p e iously applied o
S. au eus
bac e emia)11, o , in
173 CRBSI cases, absence o ca he e emo al wi hin 48 hou s. Da a quali y was moni o ed, wi h que ies
174 sen o local in es iga o s o inconsis encies o missing da a.
175
176
Da a analysis
177 Ca ego ical a iables we e p esen ed as coun s/pe cen ages, and con inuous a iables as
178 median wi h in e qua ile ange (IQR). Ca ego ical a iables we e compa ed using Chi-squa e o
179 Fishe exac es s, and con inuous a iables wi h S uden ’s - es s o Mann-Whi ney U es s. S a i ied
180 analyses we e pe o med o accoun o po en ial con ounde s and e ec modi ie s.
181 Su i al cu es o pa ien s wi h complica ed and uncomplica ed bac e emia we e plo ed and
182 compa ed using he log- ank es . Pa ien s wi hou a 30-day assessmen we e censo ed a hei las
183 ollow-up. Independen p edic o s o 30-day mo ali y we e iden i ied using Cox eg ession, wi h
184 a iables selec ed h ough a s epwise backwa d p ocedu e, including hose wi h a p- alue <0.1 in
185 uni a ia e analysis and hose o clinical o li e a u e-based ele ance. (Me hods [3], Supplemen a y
186 Ma e ial). In e ac ions we e e alua ed.
187 To assess he impac o complica ed bac e emia on mo ali y, we de eloped wo models: one
188 including he composi e a iable "complica ed bac e emia" and ano he including ins ead each
189 c i e ion o complica ed bac e emia indi idually. To assess he impac o complica ed bac e emia on
190 mo ali y, wo models we e de eloped: one wi h a composi e a iable o complica ed bac e emia and
191 ano he analyzing each c i e ion indi idually.
192 Sensi i i y analyses we e pe o med using logis ic eg ession o es he obus ness o indings,
193 accoun ing o 21 pa ien s censo ed be o e day 30. Assump ions included all 21 su i ing and andom
Page 47 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
12
194 impu a ion o wo dea hs in 10 i e a ions (Me hods [4], Supplemen a y ma e ial). Addi ionally, we
195 epea ed he mul i a iable models excluding pa ien s classi ied as ha ing a apidly a al unde lying
196 disease acco ding o he McCabe classi ica ion (i.e., conside ed o ha e a high p obabili y o dying
197 wi hin he nex h ee mon hs) o accoun o cases whe e a e minal condi ion migh p eclude ce ain
198 aspec s o managemen and in whom dea h would likely be p ima ily a ibu able o he unde lying
199 condi ion (Table S5). S a is ical analyses we e conduc ed using SPSS e sion 29.0.1.0 (SPSS Inc.,
200 Chicago, IL, USA).
201
202 RESULTS
203
204
Cha ac e is ics o he s udy coho
205 The PROBAC coho included 6,313 episodes o bac e emia, wi h CoNS isola ed om blood
206 cul u es and no conside ed con aminan s in 481 cases. A e excluding 36 cases (28 polymic obial
207 and 8 due o
S. lugdunensis
), 445 pa ien s (7.0% o he coho ) we e included in he analysis (Table
208 1).
209 Among hese pa ien s, he majo i y (268/445, 60.2%) we e male, wi h a median age o 67 yea s
210 (IQR 54-76). The mos common sou ce o in ec ion was ca he e - ela ed (336/445, 75.5%). A onse ,
211 109/445 pa ien s (24.5%) had a SOFA sco e ≥2, and 327/445 isola es (73.5%) we e me hicillin-
212 esis an . FUBC Follow-up blood cul u es we e pe o med in 122/445 pa ien s (27.4%), e ealing
213 pe sis en bac e emia in 44 o hemse cases (36.1% o pa ien s wi h FUBCs), which ep esen s 9.8%
214 o he o al coho . Compa a i e analysis be ween pa ien s wi h and wi hou FUBC (Table S1) showed
215 ha pa ien s wi h endoca di is (9.8% s. 2.2%, espec i ely; p<0.001), ca diac p os he ic al es (7.4%
216 s. 2.2%, p=0.008), and pacemake s (6.6% s. 1.5%, p=0.005) mo e equen ly had FUBC. Ac i e
Page 48 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
13
217 empi ic he apy was adminis e ed o 155 pa ien s (34.8%). Empi ic he apy was adminis e ed in 67.2%
218 o cases (299/445), and i was ac i e
in i o
in 151 (33.9%).
219 In-hospi al mo ali y was 14.8% (66/445), while 30-day mo ali y was 17.3% (77/445). Da a
220 o 21/445 pa ien s (4.7%) who we e discha ged ali e and we e no assessed a day 30 be o e day 30
221 we e censo ed a hei las ollow-up da e..
222
223
Complica ed and uncomplica ed CoNS BSI
224 Among he 445 pa ien s, 240 (53.9%) me he c i e ia o complica ed BSI (Table 2). The mos
225 common issue was delayed ca he e emo al in CRBSI (141/240 pa ien s, 58.8%), ollowed by
226 pe sis en e e o ≥72 hou s (77/240 pa ien s, 32.1%) and he p esence o a p os he ic de ice (64/240
227 pa ien s, 26.7%).
228 Pa ien s wi h complica ed BSI had a highe p e alence o mode a e- o-se e e CKD (13.3% s.
229 7.3%, p=0.045) and we e mo e likely o be on immunosupp essi e he apy (18.3% s. 10.2%,
230 p=0.016). Sep ic shock was also mo e equen (10% s. 4.4%, p=0.024). Ca he e - ela ed in ec ions
231 we e mo e p e alen in complica ed cases (80.8% s. 69.3%, p=0.005), while an unknown bac e emia
232 sou ce was mo e common in uncomplica ed cases (18.1% s. 5%, p<0.001). Addi ionally, oxacillin
233 esis ance was highe among hose wi h complica ed BSI (80.4% s. 65.4%, p<0.001).
234 Mo ali y was signi ican ly highe in he complica ed g oup (22.1% s. 11.7%, p=0.004). The
235 p opo ion o censo ed pa ien s was simila be ween g oups (4.5% s. 4.8%, p=0.88). Pa ien s wi h
236 complica ed BSI had a longe hospi al s ay (median 17 days [IQR 7-34] s. 11 days [IQR 7-23],
237 p=0.07), bu he e we e no signi ican di e ences in ICU admission o ecu ence. Su i al cu es a e
238 shown in Figu e 1.
239
Page 49 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
14
240
P edic o s o mo ali y in CoNS BSI cases
241 Bi a ia e analysis esul s a e in Table 3, and mul i a ia e esul s a e in Table 4. Iin mul i a ia e
242 model 1 (Table 4), which conside ed “complica ed BSI” as a single en i y, inc eased mo ali y was
243 associa ed wi h age (HR 1.03 pe yea , 95% CI 1.01-1.05), ce eb o ascula disease (HR 2.58, 95% CI
244 1.45-4.58), immunosupp essi e he apy (HR 2.16, 95% CI 1.22-3.84), SOFA sco e (HR 1.09 pe sco e
245 poin , 95% CI 1.03-1.16), and complica ed bac e emia (HR 2.14, 95% CI 1.29-3.53). Con e sely, a
246 ca he e sou ce o bac e emia was p o ec i e (HR 0.49, 95% CI 0.30-0.80) (Table 4). The AUROC o
247 he model was 0.75 (95%CI 0.70 – 0.81).
248 In models 2, he speci ic c i e ia o complica ed bac e emia we e included sepa a ely ins ead
249 o he composi e a iable. In model 2a, which included all pa ien s e alua ed each c i e ion o
250 “complica ed BSI” sepa a ely, ca he e emo al was ca ego ized as CRBSI wi hou ea ly ca he e
251 emo al ( e e ence), CRBSI wi h ea ly ca he e emo al, and non-CRBSI; , pe sis en e e (HR 2.52,
252 95% CI 1.52-4.17) was a isk ac o o mo ali y, while ea ly ca he e emo al in CRBSI was
253 p o ec i e (HR 0.47, 95% CI 0.26-0.83). The ca he e - ela ed o igin was no linked o mo ali y in his
254 model.In Model 2b, which included only he CRBSI cases, he same componen s we e ound o be
255 associa ed wi h mo ali y, wi h simila haza d a ios. The AUROC o models 2a and 2b we e 0.75
256 (95%CI 0.69–0.81) and 0.75 (95%CI 0.67–0.82), espec i ely.
257 Tables S21 and S32 show he bi a ia e analysis o in-hospi al mo ali y and he mul i a ia e
258 logis ic eg ession models. P edic o s o in-hospi al mo ali y we e consis en wi h hose o 30-day
259 mo ali y. Sensi i i y analysis, which impu ed wo dea hs o pa ien s discha ged ali e be o e day 30,
260 e ealed adjus ed odds a ios (aORs) o complica ed bac e emia anging om 1.90 o 2.06, wi h all
261 95% CIs abo e 1 (Table S43). Addi ionally, in he sensi i i y analysis excluding he 28 pa ien s wi h
262 a apidly a al unde lying disease, complica ed bac e emia emained associa ed wi h inc eased haza ds
Page 50 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
15
263 o mo ali y (HR 2.34, 95%CI 1.37–3.99), while ca he e emo al wi h lowe mo ali y (HR 0.46,
264 95%CI 0.25–0.85) (Table S5).
265
266 DISCUSSION
267
268 This p ospec i e coho s udy analyzed pa ien demog aphics, clinical ea u es, complica ions,
269 and ou comes o CoNS BSI. A de ini ion o complica ed bac e emia was explo ed, consis en ly linked
270 oassocia ed wi h highe 30-day mo ali y, wi h wo o he de ining c i e ia showing indi idual
271 associa ion: pe sis en e e and delayed ca he e emo al.
272 To ou knowledge, his is he la ges coho ocused solely on CoNS BSI, unde sco ing he
273 ela i e lack o esea ch compa ed o he ex ensi e li e a u e on
S. au eus
BSI. The pa ien
274 cha ac e is ics align wi h hose ypical o ca he e - ela ed BSI, such as ad anced age and high
275 como bidi y a es. No ably, abou 25% o pa ien s had sepsis a onse . The 30-day all-cause mo ali y
276 a e was 17.3%, wi h p e ious s udies epo ing a wide ange o mo ali y a es om 4% o 24% 2–7,
277 likely due o a ia ions in pa ien popula ions and c i e ia o de ining CoNS bac e emia.
278 P e ious s udies ha e iden i ied isk ac o s o mo ali y including age, lack o ac i e an ibio ic
279 ea men , and como bidi ies like enal o li e disease. 2–5 In ou s udy, ad anced age, ce eb o ascula
280 disease, and immunosupp essi e he apy we e linked o highe mo ali y, likely e lec ing pa ien
281 ail y, while he SOFA sco e indica ed acu e se e i y. Addi ionally, ca he e - ela ed in ec ions we e
282 associa ed wi h lowe mo ali y compa ed o in ec ions om o he sou ces.2–5
283 In he absence o speci ic c i e ia o CoNS BSI, ou s udy used modi ied c i e ia o iginally
284 designed o
S. au eus
complica ed bac e emia. The inding ha complica ed bac e emia is
Page 51 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
16
285 independen ly associa ed wi h highe mo ali y sugges s ha hese c i e ia migh be use ul o CoNS
286 BSI as well. Howe e , he sepa a e analysis in he second model, whe e no all c i e ia we e linked o
287 mo ali y, indica es he need o a ailo ed app oach o CoNS BSI in u u e esea ch.
288 Follow-up blood cul u esFUBC we e pe o med in ewe han 30% o cases in ou coho .
289 Al hough speci ic guidelines a e lacking, ob aining hem in conside ing ollow-up blood cul u es o
290 high- isk pa ien s wi h complica ed bac e emia may be bene icial, especially gi en he obse ed
291 associa ion be ween mo ali y and pe sis en e e , which could indica e ongoing bac e emia. In
292 con as , FUBC ollow-up blood cul u es may be unnecessa y o low- isk cases, as sugges ed by
293 Badia-Cebada e al.23 Thei ial compa ing an ibio ics e sus no an ibio ics in low- isk CoNS CRBSI
294 ound only one case o pe sis en bac e emia in he no-an ibio ics g oup. Howe e , he small sample
295 size due o s udy in e up ion limi s he s eng h o hese indings.
296 Al hough ou s udy ound ea ly ca he e emo al o be p o ec i e agains mo ali y, i was
297 ca ied ou in ewe han 60% o CRBSI cases. IDSA guidelines conside he possibili y o ca he e
298 e en ion in uncomplica ed CoNS CRBSI cases.10 Howe e , despi e he ac ha some s udies sugges
299 ha ca he e e en ion in CoNS bloods eam in ec ions may no signi ican ly impac mo ali y, i
300 emains an independen isk ac o o ecu ence,24 while ecu ence is a e a e ca he e emo al.25
301 Ou inding ha ea ly ca he e emo al is independen ly associa ed wi h a p o ec i e e ec on 30-day
302 mo ali y sugges s he need o u he s udies o be e iden i y he subg oups o pa ien s in whom
Page 52 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
17
303 e aining he ca he e is sa e. This indica es poo adhe ence o CRBSI managemen s anda ds, e en in
304 hospi als wi h dedica ed bac e emia p og ams, whe e ea ly ca he e emo al is c ucial. In cases wi hou
305 isk ac o s o complica ions, emo ing he in ec ed ca he e alone migh be su icien and could
306 po en ially educe he need o an ibio ics. Howe e , suppo ing da a o his app oach emains limi ed
307 In ou s udy, abou one- hi d o pa ien s did no ecei e empi ic he apy a he onse o in ec ion
308 onse ; howe e , hough app oxima ely 90% o hese pa ien s we e s a ed ini ia ed on on a ge ed
309 he apy once cul u e esul s we e became a ailable. Omi ing empi ic he apy may be jus i ied i no
310 isk ac o s a e p esen and sou ce con ol (e.g. ca he e emo al) is achie ed, po en ially making
311 subsequen a ge ed he apy unnecessa y , as p e iously men ioned. 23,2625–27 Howe e , empi ic he apy
312 should be ini ia ed i isk ac o s o ad e se ou comes a e iden i ied. These indings highligh he need
313 o ca e ul ini ial assessmen and isk s a i ica ion.
314 Addi ionally, nea ly hal o he empi ic he apies we e ine ec i e agains he isola ed s ain,
315 unde sco ing he impo ance o s a ing ea men ha co e s me hicillin- esis an s ains when CoNS
316 in ec ions a e suspec ed. This app oach is suppo ed by ou inding ha o e 70% o isola es in ou
317 coho we e me hicillin- esis an , consis en wi h o he epo s.
318 Ou s udy has limi a ions. Fi s , he absence o a compa a i e g oup es ic s ou abili y o
319 es ablish causal ela ionships and assess he p opo ion o mo ali y di ec ly a ibu able o CoNS BSI
320 e sus unde lying condi ions. Al hough a mino i y o pa ien s did no comple e he 30-day ollow-up,
321 sensi i i y analyses sugges he esul s a e obus . We acknowledge ha posi i e FUBC esul s likely
Page 53 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
18
322 ep esen only a subse o pe sis en bac e emia cases, as some pa ien s wi hou FUBC may ha e had
323 unde ec ed posi i e esul s; al hough we e ospec i ely analyzed ac o s in luencing he decision o
324 pe o m FUBC, he da a lacks clea indica ions o each case. None heless, we belie e including his
325 a iable is impo an o highligh he need o u he s udies o cla i y FUBC indica ions in CoNS
326 bloods eam in ec ions. Managemen o pa ien s was pe o med acco ding o a ending physicians and
327 he e o e wasa no s anda dized; none heless, we belie e he da a e lec eal-wo ld clinical p ac ice.
328 Despi e e o s like in es iga o aining and da a moni o ing, some da a collec ion e o s may s ill
329 exis . Howe e , he s udy's s eng hs include a la ge sample size, and a mul icen e , p ospec i e design
330 conduc ed by in ec ious disease expe s in hospi als wi h bac e emia p og ams, which enhances he
331 c edibili y o ou indings.
332 In summa y, ou s udy iden i ies key mo ali y ac o s in CoNS BSI and emphasizes he need
333 o imp o ed isk s a i ica ion. The p oposed de ini ion o complica ed CoNS BSI needs alida ion
334 in u he s udies.
335
336 AUTHOR CONTRIBUTIONS
337
338 Concep ualiza ion: LELC, JRB. Me hodology and s a is ical analysis: BV, LELC, JRB. O iginal d a :
339 BV. Re iew & edi ing: BV, LELC, JRB (all au ho s). Supe ision: JRB. P ojec adminis a ion:
340 LELC. Funding acquisi ion: LELC, JRB. All au ho s con ibu ed o he inal manusc ip .
341
342 FUNDING
343
Page 54 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
19
344 This esea ch was suppo ed by g an s om Plan Nacional de I+D+i 2013-2016, Ins i u o de Salud
345 Ca los III, Subdi ección Gene al de Redes y Cen os de In es igación Coope a i a, Minis e io de
346 Ciencia, Inno ación y Uni e sidades (PI16/01432); he Spanish Ne wo k o Resea ch in In ec ious
347 Diseases (REIPI) (RD16/0016/0001; RD16/ 0016/0008); and co- inanced by he Eu opean Regional
348 De elopmen Fund unde he 'A way o achie e Eu ope' Ope a ional P og am o In elligen G ow h
349 2014-2020.
350
351 TRANSPARENCY STATEMENT
352
353 PRG has se ed as a scien i ic ad iso o Shionogi and Ad anz, speake o Angelini, and Mena ini.
354 LELC has se ed as a scien i ic ad iso o No a is, speake o MSD, P ize , Angelini, and ViiV,
355 and aine o MSD and ViiV. The o he au ho s decla e no con lic s o in e es .
Page 55 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
20
356 REFERENCES
357 1. Heilmann C, Ziebuh W, Becke K. A e coagulase-nega i e s aphylococci i ulen ? Clinical
358 Mic obiology and In ec ion. 2019;25(9):1071-1080. doi:10.1016/j.cmi.2018.11.012
359
360 2. Molina J, Peñuela I, Lepe JA, e al. Mo ali y and hospi al s ay ela ed o coagulase-nega i e
361 S aphylococci bac e emia in non-c i ical pa ien s. Jou nal o In ec ion. 2013;66(2):155-162.
362 doi:10.1016/j.jin .2012.10.021
363
364 3. Pa k SY, Kwon KH, Chung JW, Huh HJ, Chae SL. Coagulase-nega i e s aphylococcal
365 bac e emia: isk ac o s o mo ali y and impac o ini ial app op ia e an imic obial he apy
366 on ou come. Eu opean Jou nal o Clinical Mic obiology and In ec ious Diseases.
367 2015;34(7):1395-1401. doi:10.1007/s10096-015-2364-3
368
369 4. Hen zien M, S ady C, Ve ne -Ga nie V, e al. P ognos ic ac o s associa ed wi h 30-day in-
370 hospi al mo ali y in coagulase-nega i e S aphylococcus bac e aemia: no impac o
371 ancomycin minimum inhibi o y concen a ion. In ec Dis. 2017;49(9):664-673.
372 doi:10.1080/23744235.2017.1323346
373
374 5. Cui J, Liang Z, Mo Z, Zhang J. The species dis ibu ion, an imic obial esis ance and isk
375 ac o s o poo ou come o coagulase-nega i e s aphylococci bac e aemia in China.
376 An imic ob Resis In ec Con ol. 2019;8(1):1-10. doi:10.1186/s13756-019-0523-5
377
378 6. Yamada K, Namikawa H, Fujimo o H, e al. Clinical cha ac e is ics o me hicillin- esis an
379 coagulase-nega i e s aphylococcal bac e emia in a e ia y hospi al. In e nal Medicine.
380 2017;56(7):781-785. doi:10.2169/in e nalmedicine.56.7715
381
382 7. Rosa RG, Dos San os RP, Goldani LZ. Mo ali y ela ed o coagulase-nega i e s aphylococcal
383 bac e emia in eb ile neu openia: A coho s udy. Canadian Jou nal o In ec ious Diseases
384 and Medical Mic obiology. 2014;25(1):14-18. doi:10.1155/2014/702621
385
386 8. Rahkonen M, Lu inen S, Koskela M, Hau ala T. T ue bac e emias caused by coagulase
387 nega i e S aphylococcus a e di icul o dis inguish om blood cul u e con aminan s.
388 Eu opean Jou nal o Clinical Mic obiology and In ec ious Diseases. 2012;31(10):2639-2644.
389 doi:10.1007/s10096-012-1607-9
390
391 9. Mi e S, Weins ein MP, Reime LG, Wilson ML, Relle LB. Rele ance o he numbe o
392 posi i e bo les in de e mining clinical signi icance o coagulase-nega i e s aphylococci in
393 blood cul u es. J Clin Mic obiol. 2001;39(9):3279-3281. doi:10.1128/JCM.39.9.3279-
394 3281.2001
395
396 10. Me mel LA, Allon M, Bouza E, e al. Clinical p ac ice guidelines o he diagnosis and
397 managemen o in a ascula ca he e - ela ed in ec ion: 2009 upda e by he In ec ious
398 Diseases Socie y o Ame ica. Clinical In ec ious Diseases. 2009;49(1):1-45.
399 doi:10.1086/599376
400
401 11. Jung N, Rieg S. Essen ials in he managemen o S. au eus bloods eam in ec ion. In ec ion.
402 2018;46(4):441-442. doi:10.1007/s15010-018-1130-8
403
Page 56 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
Table 1. Baseline pa ien and in ec ion cha ac e is ics wi h bi a ia e analysis o complica ed e sus uncomplica ed
bac e emia.
Va iables
N (%) excep whe e speci ied
All pa ien s
N = 445
Uncomplica ed
N = 205 (46.1%)
Complica ed
N = 240 (53.9%)
p
- alue1
Age, median (IQR)
67 (54 - 76)
67 (53 – 76)
67 (54 – 76)
0.950
Male gende
268 (60.2)
124 (60.5)
144 (60)
0.766
Age-adjus ed Cha lson, median (IQR)
4 (2 - 6)
4 (2 - 6)
4 (2 - 7)
0.281
Age-adjus ed Cha lson >3
261 (58.7)
118 (57.6)
143 (59.6)
0.666
Como bidi ies and condi ions
Diabe es melli us
65 (14.6)
31 (15.1)
34 (14.2)
0.776
Ch onic li e disease
39 (8.8)
23 (11.2)
16 (6.7)
0.177
Ch onic kidney disease, mode a e-se e e
47 (10.6)
15 (7.3)
32 (13.3)
0.045
Hemodialysis
9 (2.0)
1 (0.5)
8 (3.3)
0.077
Ce eb o ascula disease
50 (11.2)
25 (12.2)
25 (10.4)
0.554
COPD
40 (9.0)
18 (8.8)
22 (9.2)
0.887
Ch onic hea ailu e
37 (8.3)
16 (7.8)
21 (8.8)
0.719
Solid cance
115 (25.8)
46 (22.4)
69 (28.8)
0.130
Immunosupp essi e he apy
65 (14.6)
21 (10.2)
44 (18.3)
0.016
Recen exposu es (p e ious mon h)
An ibio ic he apy
239 (53.7)
106 (51.7)
133 (55.4)
0.434
Su ge y
124 (27.9)
61 (29.8)
63 (26.3)
0.411
Pa en e al nu i ion
97 (21.8)
48 (23.4)
49 (20.4)
0.445
In ec ion acquisi ion
Communi y
12 (2.7)
5 (2.4)
7 (2.9)
0.757
Heal hca e-associa ed
69 (15.5)
28 (13.7)
41 (17)
0.320
Nosocomial
364 (81.8)
171 (83.4)
193 (80.4)
0.354
P e ious leng h o hospi aliza ion in days,
median (IQR)
15 (9 - 27)
12 (5 – 21)
12 (3 – 24)
0.891
Hospi al depa men o admission a onse
Eme gency
3 (0.7)
2 (1)
1 (0.4)
0.473
In ensi e ca e
58 (13)
24 (11.7)
34 (14.2)
0.442
Page 63 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
Su ge y
123 (27.6)
69 (33.7)
54 (22.5)
0.009
Medicine
241 (54.2)
100 (48.8)
141 (58.8)
0.035
O he s
20 (4.5)
10 (4.9)
10 (4.2)
0.718
Vascula access de ices
PVC
229 (51.5)
113 (55.1)
116 (48.3)
0.153
Sho - e m CVC
229 (51.5)
102 (49.8)
127 (52.9)
0.506
PICC
74 (16.6)
27 (13.2)
47 (19.6)
0.070
Tunneled CVC
12 (2.7)
0
12 (5)
-
Po -a-Ca h
6 (1.4)
0
6 (2.5)
-
O he de ices
Pacemake
13 (2.9)
0
13 (5.4)
-
Vascula p os he ic de ice
9 (2)
0
9 (3.8)
-
P os he ic join
15 (3.4)
0
15 (6.3)
-
Hea al ula p os hesis
16 (3.6)
0
16 (6.7)
-
Se e i y a onse
SOFA sco e, median (IQR)
3 (1 - 6)
3 (1 - 5)
3 (1 - 6)
0.533
SOFA sco e ≥2
109 (24.5)
43 (21)
66 (27.5)
0.337
Pi sco e, median (IQR)
0 (0-3)
1 (0 - 2)
1 (0 - 3)
0.828
Pi sco e >3
87 (19.6)
39 (19)
48 (20)
0.815
Vasop esso s suppo
27 (6.1)
8 (3.9)
19 (7.9)
0.087
Sep ic shock
33 (7.4)
9 (4.4)
24 (10)
0.024
Sou ce o in ec ion
Ca he e - ela ed
336 (75.5)
142 (69.3)
194 (80.8)
0.005
Unknown
49 (11.0)
37 (18.1)
12 (5)
<0.001
Endoca di is
19 (4.3)
0
19 (7.9)
-
Skin and so issues
14 (3.2)
9 (4.4)
5 (2.1)
0.165
Abdominal
7 (1.6)
4 (2)
3 (1.3)
0.533
Bone and join s
6 (1.4)
4 (2)
2 (0.8)
0.308
O he s
14 (3.2)
9 (4.4)
5 (2.1)
0.264
CoNS species
S aphylococcus epide midis
332 (74.6)
145 (70.7)
187 (77.9)
0.083
S aphylococcus hominis
60 (13.5)
32 (15.6)
28 (11.7)
0.225
Page 64 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
S aphylococcus haemoly icus
30 (6.7)
16 (7.8)
14 (5.8)
0.408
S aphylococcus capi is
16 (3.6)
8 (3.9)
8 (3.3)
0.748
S aphylococcus wa ne i
3 (0.7)
1 (0.5)
2 (0.8)
0.657
S aphylococcus schlei e i
2 (0.5)
1 (0.5)
1 (0.4)
0.279
S aphylococcus simulans
1 (0.2)
1 (0.5)
0
0.911
S aphylococcus au icula is
1 (0.2)
1 (0.5)
0
0.279
Oxacillin esis ance
327 (73.5)
134 (65.4)
193 (80.4)
<0.001
Follow-up blood cul u es pe o med2
122 (27.4)
56 (27.3)
66 (27.5)
0.873
Posi i e ollow-up blood cul u es among hose
pe o med
44/122 (36.1)
0/56 (0)
44/66 (66.7)
-
Sou ce con ol (only CRBSI)
Ea ly (<48 h) ca he e emo al
195/336 (58)
142/142 (100)
53/194 (27.3)
-
Days un il ca he e emo al, median (IQR)
3 (2 – 8)
1 (0 – 3)
2 (2 – 7)
0.132
An imic obial ea men
Empi ic an ibio ic he apy, yes
299 (67.2)
134 (65.4)
165 (68.8)
0.449
Ac i e empi ic he apy, yes
151/299 (50.5)
76/134 (56.7)
75/165 (45.4)
0.060
Empi ic an ibio ic he apy, no
146 (32.8)
71 (34.6)
75 (31.3)
0.714
Empi ic no, a ge ed yes
129/146 (88.4)
59/71 (83.1)
70/75 (93.3)
0.054
Du a ion in days, median (IQR)
10 (7-17)
10 (6 – 16)
12 (7 – 18)
0.002
Ou comes
30-day mo ali y
77 (17.3)
24 (11.7)
53 (22.1)
0.004
ICU admission
90 (20.2)
40 (19.5)
50 (20.8)
0.999
Leng h o hospi al s ay a e BSI in days,
median (IQR)
13 (7 – 28)
11 (7 – 23)
17 (7 – 34)
0.079
Recu ence
8 (1.8)
3 (1.5)
5 (2.1)
0.459
COPD: ch onic obs uc i e pulmona y disease; PVC: pe iphe al enous ca he e ; CVC: cen al
enous ca he e ; PICC: pe iphe ally inse ed cen al ca he e ; CRBSI: ca he e - ela ed bloods eam
in ec ion; CoNS: coagulase-nega i e s aphylococci; ICU: in ensi e ca e uni ; BSI: bloods eam
in ec ion
Page 65 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
1 S a is ical analyses we e no pe o med o a iables ha a e c i e ia o complica ed bac e emia;
hey a e included solely o p o ide da a o uncomplica ed bac e emia cases.
2 Follow-up blood cul u es we e pe o med a he disc e ion o he a ending physician, wi h a
speci ic analysis o pa ien cha ac e is ics associa ed wi h ob aining hem p o ided in he
Supplemen a y Ma e ial.
Page 66 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
Table 2. Complica ed bac e emia. C i e ia o complica ed bac e emia among he 240 pa ien s wi h
episodes classi ied as such. Pa ien s may ha e had mo e han one c i e ion.
C i e ia o complica ed bac e emia
N (%)
Absence o ea ly (<48h) ca he e emo al in CRSBI
141 (58.8)
Pe sis en e e ≥72h
77 (32.1)
Implan ed de ice
64 (26.7)
Pe sis en bac e emia
44 (18.3)
Me as a ic complica ions
7 (2.9)
P ima y endoca di is
19 (7.9)
Page 67 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
Table 2. Complica ed bac e emia. C i e ia o complica ed bac e emia among he 240 pa ien s wi h
episodes classi ied as such. Pa ien s may ha e had mo e han one c i e ion.
C i e ia o complica ed bac e emia
N (%)
Absence o ea ly (<48h) ca he e emo al in CRSBI
141 (58.8)
Pe sis en e e ≥72h
77 (32.1)
Implan ed de ice
64 (26.7)
Pe sis en bac e emia
44 (18.3)
Me as a ic complica ions
7 (2.9)
P ima y endoca di is
19 (7.9)
Page 68 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
Table 3. Bi a ia e analysis o he associa ion o di e en a iables wi h 30-day mo ali y.
Va iables
N (%) excep whe e speci ied
Ali e
N = 368
(82.7%)
Dead
N = 77
(17.3%)
HR
95%CI
p
- alue
Age, median (IQR)
66 (53 – 75)
75 (62 – 81)
1.03
1.01 – 1.04
0.001
Male gende
222 (60.3)
46 (59.7)
1.02
0.64 – 1.61
0.943
Age-adjus ed Cha lson, median (IQR)
4 (2 - 6)
2 (1 - 5)
1.09
1.01 – 1.18
0.039
Age-adjus ed Cha lson >3
208 (56.5)
53 (68.8)
1.62
0.99 – 2.65
0.053
Como bidi ies and condi ions
Diabe es melli us
55 (15)
10 (13)
0.83
0.43 – 1.61
0.575
Ch onic li e disease
33 (9)
6 (7.8)
0.86
0.38 – 1.99
0.732
Ch onic kidney disease, mode a e-se e e
34 (9.2)
13 (16.9)
1.94
1.07 – 3.53
0.029
Hemodialysis
7 (1.9)
2 (2.6)
1.13
0.26 – 4.79
0.874
Ce eb o ascula disease
34 (9.2)
16 (20.8)
2.22
1.28 – 3.85
0.005
COPD
31 (8.4)
9 (11.7)
1.42
0.71 – 2.84
0.326
Ch onic hea ailu e
25 (6.8)
12 (15.6)
2.25
1.21 – 4.16
0.010
Solid cance
91 (24.7)
24 (31.2)
1.31
0.81 – 2.13
0.269
Immunosupp essi e he apy
48 (13)
17 (22.1)
1.70
0.99 – 2.92
0.054
Recen exposu es (p e ious mon h)
An ibio ic he apy
191 (51.9)
48 (62.3)
1.37
0.86 – 2.18
0.188
Su ge y
107 (29.1)
17 (22.1)
0.68
0.39 – 1.17
0.167
Pa en e al nu i ion
82 (22.3)
15 (19.5)
0.82
0.47 – 1.45
0.497
In ec ion acquisi ion
Communi y
11 (3)
1 (1.3)
0.45
0.06 – 3.21
0.422
Heal hca e-associa ed
51 (13.9)
18 (23.4)
1.80
1.06 – 3.06
0.029
Nosocomial
305 (82.9)
58 (75.3)
0.65
0.39 – 1.09
0.107
P e ious leng h o hospi aliza ion in days,
median (IQR)
12 (5 – 23)
11 (3 – 24)
1.01
0.99 – 1.01
0.492
Hospi al depa men o admission a onse
Eme gency
3 (0.8)
0
-
-
0.606
In ensi e ca e
41 (11.1)
17 (22.1)
2.03
1.18 – 3.48
0.010
Page 69 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
Su ge y
111 (30.2)
12 (15.6)
0.45
0.25 – 0.84
0.012
Medicine
195 (53)
46 (59.7)
1.31
0.83 – 2.08
0.246
O he s
18 (4.9)
2 (2.6)
0.35
0.05 – 2.51
0.295
Vascula access de ices
PVC
184 (50)
45 (58.4)
1.33
0.84 – 2.10
0.223
Sho - e m CVC
191 (51.9)
38 (49.4)
0.87
0.55 – 1.36
0.530
PICC
63 (17.1)
11 (14.3)
0.79
0.42 – 1.51
0.490
Tunneled CVC
8 (2.2)
4 (5.2)
2.22
0.81 – 6.09
0.119
Po -a-Ca h
4 (1.1)
3 (3.9)
3.15
0.99 – 10.01
0.051
O he de ices
Pacemake
8 (2.2)
5 (6.5)
3.18
1.29 – 7.89
0.012
Vascula p os he ic de ice
6 (1.6)
3 (3.9)
2.26
0.71 – 7.16
0.168
P os he ic join
11 (3)
4 (5.2)
1.60
0.59 – 4.38
0.359
Hea al ula p os hesis
11 (3)
5 (6.5)
2.05
0.83 – 5.07
0.122
Se e i y a onse
SOFA sco e, median (IQR)
3 (1 - 5)
4 (1 - 8)
1.08
1.01 – 1.14
0.015
SOFA sco e ≥2
80 (21.7)
29 (37.7)
1.96
1.24 – 3.12
0.004
Pi sco e, median (IQR)
1 (0 - 3)
0 (0 – 2)
0.99
0.91 – 1.09
0.893
Pi sco e >3
74 (20.1)
13 (16.9)
0.86
0.47 – 1.56
0.619
Vasop esso s suppo
16 (4.4)
11 (14.3)
2.29
0.97 – 5.39
0.059
Sep ic shock
16 (4.4)
17 (22.1)
5.09
2.96 – 8.75
<0.001
Sou ce o in ec ion
Ca he e - ela ed
287 (78)
49 (63.6)
0.472
0.29 – 0.75
0.002
Unknown
40 (10.9)
9 (11.7)
1.38
0.69 – 2.76
0.369
P ima y eEndoca di is
13 (3.5)
6 (7.8)
2.03
0.88 – 4.66
0.097
Skin and so issues
10 (2.7)
4 (5.2)
1.86
0.68 – 5.08
0.228
Abdominal
6 (1.6)
1 (1.3)
0.78
0.11 – 5.61
0.805
Bone and join s
5 (1.4)
1 (1.3)
0.99
0.14 – 7.18
0.999
O he s
1 (0.3)
0
-
-
0.766
CoNS species
S aphylococcus epide midis
278 (75.5)
54 (70.1)
0.75
0.46 – 1.22
0.246
S aphylococcus hominis
48 (13)
12 (15.6)
1.21
0.65 – 2.24
0.545
Page 70 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
S aphylococcus haemoly icus
22 (6)
8 (10.4)
1.93
0.93 – 4.01
0.080
S aphylococcus capi is
14 (3.8)
2 (2.6)
0.72
0.18 – 2.95
0.652
S aphylococcus wa ne i
2 (0.5)
1 (1.3)
1.89
0.26 – 13.65
0.525
S aphylococcus schlei e i
2 (0.5)
0
-
-
0.674
S aphylococcus simulans
1 (0.3)
0
-
-
0.766
S aphylococcus au icula is
1 (0.3)
0
-
-
0.766
Oxacillin esis ance
266 (72.3)
61 (79.2)
1.42
0.81 – 2.49
0.228
Follow-up blood cul u es pe o med
101 (27.5)
21 (27.3)
0.99
0.59 – 1.67
0.998
Posi i e ollow-up blood cul u es among
hose pe o med
29 (7.9)
15 (19.5)
2.34
1.36 – 4.20
0.003
Sou ce con ol (only CRBSI)
Ea ly ca he e emo al
173/287 (60.3)
22/49 (44.9)
0.49
0.28 – 0.88
0.016
Days un il ca he e emo al, median
(IQR)
2 (1 – 6)
4 (1 – 8)
0.85
0.73 – 0.99
0.033
Complica ed bac e emia
187 (50.8)
53 (68.8)
2.09
1.28 – 3.40
0.003
Absence o ea ly (<48h) ca he e emo al
in CRSBI
114/287 (39.7)
27/49 (55.1)
2.04
1.14 - 3.57
0.016
Pe sis en e e ≥72h
52 (14.1)
25 (32.5)
2.48
1.54 – 4.01
<0.001
Implan ed de ice
44 (12)
20 (26)
2.44
1.47 – 4.07
<0.001
Pe sis en bac e emia
29 (7.9)
15 (19.5)
2.34
1.36 – 4.20
0.003
Me as a ic complica ions
4 (1.1)
3 (3.9)
2.39
0.76 – 7.60
0.138
An imic obial ea men
Empi ic an ibio ic he apy, yes
249 (67.7)
50 (64.9)
0.39
0.22 – 0.69
0.441
Ac i e empi ic he apy, yes
128/249 (51.4)
23/50 (46)
0.88
0.67 - 1.17
0.383
Empi ic an ibio ic he apy, no
119 (32.3)
27 (35.1)
1.21
0.75 - 194
0.549
Empi ic no, a ge ed yes
104/119 (87.4)
25/27 (92.6)
0.98
0.23 – 4.14
0.976
Du a ion in days, median (IQR)
11 (7 – 18)
8 (5 – 13)
0.96
0.92 – 0.99
0.009
Ou comes
ICU admission
66 (17.9)
24 (31.2)
1.89
1.17 – 3.07
0.010
Leng h o hospi al s ay a e BSI in days,
median (IQR)
14 (7 – 31)
12 (5 – 24)
0.99
0.99 – 1.00
0.219
Recu ence
5 (1.4)
3 (3.9)
2.39
0.75 – 7.58
0.140
Page 71 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
CKD: ch onic kidney disease; COPD: ch onic obs uc i e pulmona y disease; PICC: pe iphe ally
inse ed cen al ca he e ; PVC: pe iphe al enous ca he e ; CVC: cen al enous ca he e ;
CRBSI: ca he e - ela ed bloods eam in ec ion.
Page 72 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
Table 4. Mul i a ia e analysis o ac o s associa ed wi h 30-day mo ali y. Model 1 included
complica ed bloods eam in ec ion, and Models 2 included he indi idual componen s o
complica ed bloods eam in ec ion. In Model 2a, he “ea ly ca he e emo al” a iable was
s a i ied in o h ee ca ego ies o include all pa ien s. Model 2b included only pa ien s wi h
CRBSI.
Model 1
Va iables
HR
95% CI
p
- alue
Age
1.03
1.01 – 1.05
0.002
Ce eb o ascula disease
2.58
1.45 – 4.58
0.001
Immunosupp essi e he apy
2.16
1.22 – 3.84
0.008
SOFA sco e
1.09
1.03 – 1.16
0.005
Ca he e - ela ed bac e emia
0.49
0.30 – 0.80
0.004
Complica ed bac e emia
2.14
1.29 – 3.53
0.003
Model 2a
Va iables
HR
95% CI
p
- alue
Age
1.03
1.01 – 1.05
<0.001
Ce eb o ascula disease
2.49
1.41 – 4.39
0.002
Immunosupp essi e he apy
1.99
1.11 – 3.57
0.o020
SOFA sco e
1.09
1.03 – 1.16
0.005
Fe e ≥72 hou s
2.52
1.52 – 4.17
<0.001
Ca he managemen
CRBSI wi hou ea ly ca he e emo al
Re e ence
-
-
CRBSI wi h ea ly ca he e emo al
0.47
0.26 – 0.83
0.010
CoNS BSI o non-ca he e o igin
1.14
0.66 – 1.98
0.638
Model 2b
Page 79 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
Va iables
HR
95% CI
p
- alue
Age
1.03
1.01 – 1.06
0.003
Ce eb o ascula disease
2.92
1.49 – 5.70
0.002
Immunosupp essi e he apy
2.22
1.09 – 4.52
0.028
SOFA sco e
1.10
1.02 – 1.18
0.019
Fe e ≥72 hou s
2.20
1.14 – 4.23
0.018
Ea ly ca he e emo al
0.47
0.26 – 0.84
0.010
Page 80 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
Figu e 1. Cumula i e su i al o pa ien s wi h complica ed ( ed line) and non-complica ed (blue
line) bac e emia due o coagulase-nega i e S aphylococcus. P alue (log ank es ) = 0.004
Day
0
5
10
15
20
25
30
NC
205
193
198
180
177
174
170
C
240
220
206
198
192
186
175
Numbe a isk (NC: non complica ed; C: complica ed)
Page 81 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
1
1SUPPLEMENTARY MATERIAL
2
3
4Me hods [1]. S udy mic obiological and clinical inclusion c i e ia o CoNS bloods eam in ec ion
5episodes
6
7Episodes o CoNS BSI we e included in he s udy i hey me bo h mic obiological and clinical c i e ia as ollows:
8
91. Mic obiological c i e ia. The same CoNS species wi h iden ical an ibio ic suscep ibili y p o iles was
10 isola ed om a leas wo dis inc se s o blood cul u es d awn om sepa a e si es. In he case o
11 ca he e - ela ed CoNS BSI:
12 -I he ca he e was emo ed and cul u ed, a diagnosis o CoNS CRBSI was con i med i he same
13 CoNS species wi h an iden ical suscep ibili y p o ile was isola ed om a leas one pe iphe al blood
14 cul u e and om a cul u e o he ca he e ip, in ei he a semiquan i a i e ( oll pla e, wi h p esence
15 o 15 CFU pe pla e o mo e) o quan i a i e ( o ex wi h a coun o a leas 103 CFU/segmen , o
16 sonica ion me hods, wi h coun s abo e 102 CFU/segmen ) echniques
17 -I he ca he e was no emo ed, he same CoNS species had o be isola ed om a leas one
18 pe iphe al blood cul u e and om a sample d awn om he ca he e hub. Diagnosis equi ed ei he
19 a di e en ial ime o posi i i y ( ime o de ec ed mic obial g ow h) o a leas 120 minu es lowe in
20 he ca he e sample compa ed o he sample om pe iphe al enipunc u e; o quan i a i e blood
21 cul u es, de ined as a h ee- old highe colony coun in he blood sample ob ained om he ca he e
22 compa ed o he pe iphe al blood sample.
23
24 2. Clinical signs o in ec ion. A leas wo o he ollowing clinical signs had o be p esen :
25 -Fe e (T >38ºC) o hypo he mia (T <36ºC)
26 -Tachyca dia (>90 bea s pe minu e)
27 -Tachypnea (>20 b ea hs pe minu e) o PaCO₂ < 32 mmHg
28 -Leukocy osis (>12,000 cells/mm³) o leukopenia (<4,000 cells/mm³)
29
30 3. O he in ec ion causes we e excluded. Fo pa ien s wi h an unusual sou ce o bac e emia o CoNS
31 (i.e., pneumonia, in aabdominal in ec ion, meningi is, unknown sou ce in pa ien s wi hou ca he e s o
32 p os he ic ma e ial), he opinion o a second local in es iga o was sough .
33
Page 82 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
2
34 Me hods [2]. Jus i ica ion o he inclusion o “me as a ic complica ions” as a c i e ion o “complica ed
35 bac e emia”
36
37 Al hough me as a ic complica ions occu ed in ewe han 3% o cases, his was e ained as a c i e ion o
38 complica ed bac e emia due o he p e-de ined na u e o he de ini ion. No ably, 5 o he 7 pa ien s wi h
39 me as a ic complica ions also me o he c i e ia o "complica ed bac e emia" as ollows:
40 -1 pa ien was a ascula p os he ic de ice ca ie
41 -1 had an a icula de ice
42 -2 had pe sis en e e
43 -1 had a CRBSI wi hou ea ly ca he e emo al.
44 Only 2 pa ien s did no mee any o he "complica ed c i e ia," and he de ini ion o complica ed bac e emia in
45 hese cases was solely a ibu ed o he p esence o me as a ic oci.
46
47
Page 83 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
3
49 Me hods [3]. Mul i a iable eg ession model me hodology
50
51 In he eg ession analysis, we employed a backwa d s epwise p ocedu e. Va iables we e ini ially included based
52 on clinical ele ance (e.g., hei po en ial con ounding o modi ying e ec on he associa ion be ween
53 complica ed bac e emia and mo ali y, o hei di ec associa ion wi h mo ali y) and s a is ical signi icance (i.e.,
54 p- alue < 0.1 in bi a ia e analysis). The a iables included we e age, Cha lson como bidi y index, ch onic kidney
55 disease, ce eb o ascula disease, ch onic hea ailu e, immunosupp essi e he apy, heal hca e-associa ed
56 acquisi ion, wa d o admission, p esence o a po -a-ca h o pacemake , SOFA sco e, asop esso use, sep ic
57 shock, ca he e - ela ed sou ce, endoca di is, and complica ed bac e emia. Collinea i y was es ed, and collinea
58 a iables we e emo ed based on clinical easoning and a iance in la ion ac o (VIF) alues. Speci ically,
59 asop esso use, ICU admission a onse , and sep ic shock we e excluded due o collinea i y wi h he SOFA
60 sco e. The Cha lson como bidi y index was eplaced by indi idual como bidi ies. All a iables we e ea ed as
61 ca ego ical, excep o age and SOFA sco e, which we e ea ed as con inuous a iables.
62
63 In Model 1, he a iables ha cons i u e he c i e ia o complica ed bac e emia we e excluded, while
64 complica ed bac e emia was e ained in all s eps. Va iables we e disca ded sequen ially based on he backwa d
65 s epwise p ocedu e, elimina ing he a iable wi h he lowes in luence on model pe o mance, as de e mined by
66 he Akaike In o ma ion C i e ion (AIC). The inal model included 424 pa ien s, wi h 21 excluded due o missing
67 da a.
68
69 In Model 2, we included indi idual c i e ia o complica ed bac e emia and excluded he composi e a iable. The
70 selec ion p ocedu e o a iables was iden ical o Model 1. Model 2a included 424 pa ien s (21 missing da a),
71 conside ing "ea ly ca he e emo al" as a 3-ca ego y a iable: ca he e - ela ed bac e emia wi hou ea ly ca he e
72 emo al ( e e ence ca ego y), ca he e - ela ed bac e emia wi h ea ly ca he e emo al, and non-ca he e - ela ed
73 bac e emia. Model 2b included 324 pa ien s (12 missing alues), ocusing solely on pa ien s wi h ca he e -
74 ela ed bac e emia.
75
76 In e ac ions be ween complica ed bac e emia and he o he a iables in he inal model we e assessed, as well
77 as in e ac ions be ween he a iables and each indi idual c i e ion de ining complica ed bac e emia.
78
79 The p edic i e abili y o he inal models was assessed by calcula ing he a ea unde he ecei e ope a ing
80 cha ac e is ic (AUROC) cu e wi h 95% con idence in e als.
81
82
Page 84 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
4
83 Me hods [4]:. Sensi i i y analysis
84
85 Fo 21 pa ien s, con i ma ion o su i al a day 30 was lacking. Consequen ly, hese pa ien s we e excluded
86 om he p ima y mul i a iable analysis o mo ali y p edic o s. All hese pa ien s we e discha ged ali e. Fo
87 pa ien s wi h comple e da a, 8 ou o 379 pa ien s who we e discha ged ali e had died by day 30 (2%; 99% CI
88 0-5%). Based on his, an an icipa ion o 0-1 dea hs among he 21 pa ien s was es ablished. Subsequen ly, wo
89 scena ios we e examined:
90
91 (a) assuming all 21 pa ien s su i ed a day 30; a mul i a ia e analysis o mo ali y p edic o s was pe o med
92 using logis ic eg ession conside ing ha all 21 pa ien s we e ali e a day 30.
93
94 (b) conside ing he possibili y o up o 2 dea hs among hem ( wo- old he uppe limi o 99% CI). Dea h was
95 andomly assigned o 2 pa ien s om he 21 wi h 10 i e a ions, and mul i a ia e logis ic eg ession was
96 conduc ed o each i e a ion.
97
Page 85 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
5
98 Table S1. Fea u es o pa ien s acco ding o whe he ollow-up blood cul u es we e pe o med
99
100
FUBC pe o med
N=122
FUBC no
pe o med
N=323
OR (95% CI)
p- alue
Fe e >72h
27 (22.1)
50 (15.5)
1.55 (0.92 - 2.62)
0.098
Lack o ea ly ca he e
emo al (CRBSI only)
21/89 (23.6)
120/247 (48.6)
0.50 (0.33 - 0.77)
<0.001
Endoca di is
12 (9.8)
7 (2.2)
4.93 (1.89 - 12.82)
<0.001
Ca diac p os he ic al e
9 (7.4)
7 (2.2)
3.60 (1.31 – 9.88)
0.008
Pacemake
8 (6.6)
5 (1.5)
4.46 (1.43 – 13.92)
0.005
A icula p os he ic de ice
6 (4.9)
9 (2.8)
1.81 (0.63 – 5.18)
0.266
Endo ascula p os hesis
5 (4.1)
4 (1.2)
3.41 (0.90 – 12.91)
0.056
Me as a ic oci
3 (2.5)
4 (1.2)
2.01 (0.44 – 9.12)
0.356
SOFA sco e – median
(IQR)
3 (1 – 5)
3 (1 – 6)
0.97 (0.91 - 1.04)
0.407
101
Page 86 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
6
102
103 Table S21. Bi a ia e analysis o he associa ion o di e en exposu es wi h in-hospi al mo ali y.
104
Va iable
OR
95% CI
p- alue
Age, median (IQR)
1.03
1.01 – 1.05
<0.001
Age > 65
2.48
1.39 - 4.42
0.002
Male gende
1.26
0.74 – 2.15
0.402
Age-adjus ed Cha lson
1.07
0.90 – 1.26
0.442
Age-adjus ed Cha lson > 3
2.40
1.34 – 4.27
0.002
Diabe es melli us
1.03
0.50 – 2.14
0.936
Ch onic li e disease
0.82
0.31 – 2.17
0.683
Mode a e- o-se e e CKD
2.14
1.05 – 4.37
0.034
Hemodialysis
1.44
0.28 – 7.53
0.648
Ce eb o ascula disease
2.79
1.43 – 5.45
0.002
COPD
1.47
0.64 – 3.34
0.359
Ch onic hea ailu e
3.13
1.49 – 6.59
0.002
Rheuma ologic disease
0.51
0.07 – 3.98
1.000
Solid malignancy
1.27
0.73 – 2.29
0.416
Immunosupp essi e he apy
1.89
0.99 – 3.61
0.050
Recen an ibio ic he apy (30 days)
1.48
0.87 – 2.52
0.145
Recen su ge y (30 days)
0.53
0.27 – 1.03
0.049
Communi y-acqui ed
0.97
0.95 – 0.99
0.228
Hospi al-acqui ed
0.61
0.33 – 1.13
0.112
Heal hca e-associa ed
1.92
1.02 – 3.62
0.040
PICC
1.00
0.49 – 2.02
0.998
PVC
1.41
0.83 – 2.38
0.201
Sho - e m CVC
0.87
0.52 – 1.46
0.511
Tunneled ca he e
3.42
0.97 –
12.02
0.068
Po -a-Ca h®
0.96
0.11 – 8.06
1.000
Pacemake
2.65
0.79 – 8.85
0.112
Vascula p os he ic de ice
1.66
0.34 – 8.15
0.628
A icula p os hesis
1.45
0.39 – 5.29
0.476
Ca diac al ula p os hesis
1.97
0.62 – 6.29
0.275
Pa en e al nu i ion
0.77
0.39 – 1.50
0.437
Sep ic shock
5.07
2.40 –
10.72
<0.001
SOFA sco e
1.20
1.07 – 1.34
0.002
SOFA sco e ≥ 2
1.92
1.11 – 3.35
0.019
Pi sco e
1.26
1.12 – 1.43
<0.001
Pi sco e > 3
3.57
1.57 – 8.12
0.004
ICU admission
1.75
0.97 – 3.15
0.072
Amine suppo
2.55
0.88 – 7.40
0.080
Oxacillin esis ance
1.54
0.73 - 3.27
0.260
Ca he e - ela ed ocus
0.55
0.31 – 0.96
0.034
Unknown ocus
1.14
0.51 – 2.55
0.755
U ina y ocus
0.99
0.98 – 0.99
0.348
Skin and so issues ocus
1.59
0.43 – 5.87
0.480
Endoca di is ocus
2.82
1.03 – 7.69
0.036
Respi a o y ocus
6.05
1.47 –
24.82
0.005
Abdominal ocus
0.96
0.11 – 8.08
0.967
Bone and join s ocus
1.15
0.13 –
10.01
0.899
No empi ic he apy ecei ed
1.03
0.59 - 1.79
0.922
Adequacy o empi ic he apy
0.79
0.42 – 1.50
0.468
Du a ion empi ic, days
0.99
0.97 – 1.01
0.481
Du a ion a ge ed, days
0.99
0.99 – 1.01
0.353
Page 87 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
7
Du a ion o he apy o e all, days
0.95
0.92 – 0.99
0.013
Ca he e emo al in CRBSI
0.59
0.31 - 1.11
0.101
Pe sis en e e a 72h
3.19
1.787 –
5.71
<0.001
Pe sis en bac e emia
2.74
1.350 –
5.57
0.005
Sep ic complica ions
4.45
0.974 –
20.37
0.036
Complica ed BSI
2.19
1.25 – 3.85
0.005
Days o ca he e emo al
0.99
0.84 – 1.17
0.888
ICU s ay, days
0.99
0.98 – 1.01
0.758
BSI onse o discha ge, days
0.99
0.99 – 1.01
0.327
Admission o BSI, days
1.00
0.99 – 1.01
0.447
105 CKD: ch onic kidney disease; COPD: ch onic obs uc i e pulmona y disease; AIDS: acqui ed
106 immunode iciency synd ome; PICC: pe iphe ally inse ed cen al ca he e ; PVC: pe iphe al enous ca he e ;
107 CVC: cen al enous ca he e ; ICU: in ensi e ca e uni ; CRBSI: ca he e - ela ed bloods eam in ec ion; BSI:
108 bloods eam in ec ion
109
Page 88 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
14
(a) Gi e unadjus ed es ima es and, i applicable, con ounde -adjus ed es ima es
and hei p ecision (eg, 95% con idence in e al). Make clea which con ounde s
we e adjus ed o and why hey we e included
10-
1112-
13
(b) Repo ca ego y bounda ies when con inuous a iables we e ca ego ized
12-
1310-
11
Main esul s
16
(c) I ele an , conside ansla ing es ima es o ela i e isk in o absolu e isk o a
meaning ul ime pe iod
-
O he analyses
17
Repo o he analyses done—eg analyses o subg oups and in e ac ions, and
sensi i i y analyses
1113,
S3
Discussion
Key esul s
18
Summa ize key esul s wi h e e ence o s udy objec i es
11-
1213-
14
Limi a ions
19
Discuss limi a ions o he s udy, aking in o accoun sou ces o po en ial bias o
imp ecision. Discuss bo h di ec ion and magni ude o any po en ial bias
13-
1415
In e p e a ion
20
Gi e a cau ious o e all in e p e a ion o esul s conside ing objec i es, limi a ions,
mul iplici y o analyses, esul s om simila s udies, and o he ele an e idence
11-
1414-
16
Gene alizabili y
21
Discuss he gene alizabili y (ex e nal alidi y) o he s udy esul s
1415
O he in o ma ion
Funding
22
Gi e he sou ce o unding and he ole o he unde s o he p esen s udy and, i
applicable, o he o iginal s udy on which he p esen a icle is based
1417
198
199 *Gi e in o ma ion sepa a ely o exposed and unexposed g oups.
200
201 No e: An Explana ion and Elabo a ion a icle discusses each checklis i em and gi es me hodological
202 backg ound and published examples o anspa en epo ing. The STROBE checklis is bes used in
203 conjunc ion wi h his a icle ( eely a ailable on he Web si es o PLoS Medicine a
204 h p://www.plosmedicine.o g/, Annals o In e nal Medicine a h p://www.annals.o g/, and Epidemiology a
205 h p://www.epidem.com/). In o ma ion on he STROBE Ini ia i e is a ailable a h p://www.s obe-
206 s a emen .o g.
Page 95 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
1
1SUPPLEMENTARY MATERIAL
2
3
4Me hods [1]. S udy mic obiological and clinical inclusion c i e ia o CoNS bloods eam in ec ion
5episodes
6
7Episodes o CoNS BSI we e included in he s udy i hey me bo h mic obiological and clinical c i e ia as ollows:
8
91. Mic obiological c i e ia. The same CoNS species wi h iden ical an ibio ic suscep ibili y p o iles was
10 isola ed om a leas wo dis inc se s o blood cul u es d awn om sepa a e si es. In he case o
11 ca he e - ela ed CoNS BSI:
12 -I he ca he e was emo ed and cul u ed, a diagnosis o CoNS CRBSI was con i med i he same
13 CoNS species wi h an iden ical suscep ibili y p o ile was isola ed om a leas one pe iphe al blood
14 cul u e and om a cul u e o he ca he e ip, in ei he a semiquan i a i e ( oll pla e, wi h p esence
15 o 15 CFU pe pla e o mo e) o quan i a i e ( o ex wi h a coun o a leas 103 CFU/segmen , o
16 sonica ion me hods, wi h coun s abo e 102 CFU/segmen ) echniques
17 -I he ca he e was no emo ed, he same CoNS species had o be isola ed om a leas one
18 pe iphe al blood cul u e and om a sample d awn om he ca he e hub. Diagnosis equi ed ei he
19 a di e en ial ime o posi i i y ( ime o de ec ed mic obial g ow h) o a leas 120 minu es lowe in
20 he ca he e sample compa ed o he sample om pe iphe al enipunc u e; o quan i a i e blood
21 cul u es, de ined as a h ee- old highe colony coun in he blood sample ob ained om he ca he e
22 compa ed o he pe iphe al blood sample.
23
24 2. Clinical signs o in ec ion. A leas wo o he ollowing clinical signs had o be p esen :
25 -Fe e (T >38ºC) o hypo he mia (T <36ºC)
26 -Tachyca dia (>90 bea s pe minu e)
27 -Tachypnea (>20 b ea hs pe minu e) o PaCO₂ < 32 mmHg
28 -Leukocy osis (>12,000 cells/mm³) o leukopenia (<4,000 cells/mm³)
29
30 3. O he in ec ion causes we e excluded. Fo pa ien s wi h an unusual sou ce o bac e emia o CoNS
31 (i.e., pneumonia, in aabdominal in ec ion, meningi is, unknown sou ce in pa ien s wi hou ca he e s o
32 p os he ic ma e ial), he opinion o a second local in es iga o was sough .
33
Page 96 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
2
34 Me hods [2]. Jus i ica ion o he inclusion o “me as a ic complica ions” as a c i e ion o “complica ed
35 bac e emia”
36
37 Al hough me as a ic complica ions occu ed in ewe han 3% o cases, his was e ained as a c i e ion o
38 complica ed bac e emia due o he p e-de ined na u e o he de ini ion. No ably, 5 o he 7 pa ien s wi h
39 me as a ic complica ions also me o he c i e ia o "complica ed bac e emia" as ollows:
40 -1 pa ien was a ascula p os he ic de ice ca ie
41 -1 had an a icula de ice
42 -2 had pe sis en e e
43 -1 had a CRBSI wi hou ea ly ca he e emo al.
44 Only 2 pa ien s did no mee any o he "complica ed c i e ia," and he de ini ion o complica ed bac e emia in
45 hese cases was solely a ibu ed o he p esence o me as a ic oci.
46
47
Page 97 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
3
49 Me hods [3]. Mul i a iable eg ession model me hodology
50
51 In he eg ession analysis, we employed a backwa d s epwise p ocedu e. Va iables we e ini ially included based
52 on clinical ele ance (e.g., hei po en ial con ounding o modi ying e ec on he associa ion be ween
53 complica ed bac e emia and mo ali y, o hei di ec associa ion wi h mo ali y) and s a is ical signi icance (i.e.,
54 p- alue < 0.1 in bi a ia e analysis). The a iables included we e age, Cha lson como bidi y index, ch onic kidney
55 disease, ce eb o ascula disease, ch onic hea ailu e, immunosupp essi e he apy, heal hca e-associa ed
56 acquisi ion, wa d o admission, p esence o a po -a-ca h o pacemake , SOFA sco e, asop esso use, sep ic
57 shock, ca he e - ela ed sou ce, endoca di is, and complica ed bac e emia. Collinea i y was es ed, and collinea
58 a iables we e emo ed based on clinical easoning and a iance in la ion ac o (VIF) alues. Speci ically,
59 asop esso use, ICU admission a onse , and sep ic shock we e excluded due o collinea i y wi h he SOFA
60 sco e. The Cha lson como bidi y index was eplaced by indi idual como bidi ies. All a iables we e ea ed as
61 ca ego ical, excep o age and SOFA sco e, which we e ea ed as con inuous a iables.
62
63 In Model 1, he a iables ha cons i u e he c i e ia o complica ed bac e emia we e excluded, while
64 complica ed bac e emia was e ained in all s eps. Va iables we e disca ded sequen ially based on he backwa d
65 s epwise p ocedu e, elimina ing he a iable wi h he lowes in luence on model pe o mance, as de e mined by
66 he Akaike In o ma ion C i e ion (AIC). The inal model included 424 pa ien s, wi h 21 excluded due o missing
67 da a.
68
69 In Model 2, we included indi idual c i e ia o complica ed bac e emia and excluded he composi e a iable. The
70 selec ion p ocedu e o a iables was iden ical o Model 1. Model 2a included 424 pa ien s (21 missing da a),
71 conside ing "ea ly ca he e emo al" as a 3-ca ego y a iable: ca he e - ela ed bac e emia wi hou ea ly ca he e
72 emo al ( e e ence ca ego y), ca he e - ela ed bac e emia wi h ea ly ca he e emo al, and non-ca he e - ela ed
73 bac e emia. Model 2b included 324 pa ien s (12 missing alues), ocusing solely on pa ien s wi h ca he e -
74 ela ed bac e emia.
75
76 In e ac ions be ween complica ed bac e emia and he o he a iables in he inal model we e assessed, as well
77 as in e ac ions be ween he a iables and each indi idual c i e ion de ining complica ed bac e emia.
78
79 The p edic i e abili y o he inal models was assessed by calcula ing he a ea unde he ecei e ope a ing
80 cha ac e is ic (AUROC) cu e wi h 95% con idence in e als.
81
82
Page 98 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
4
83 Me hods [4]. Sensi i i y analysis
84
85 Fo 21 pa ien s, con i ma ion o su i al a day 30 was lacking. Consequen ly, hese pa ien s we e excluded
86 om he p ima y mul i a iable analysis o mo ali y p edic o s. All hese pa ien s we e discha ged ali e. Fo
87 pa ien s wi h comple e da a, 8 ou o 379 pa ien s who we e discha ged ali e had died by day 30 (2%; 99% CI
88 0-5%). Based on his, an an icipa ion o 0-1 dea hs among he 21 pa ien s was es ablished. Subsequen ly, wo
89 scena ios we e examined:
90
91 (a) assuming all 21 pa ien s su i ed a day 30; a mul i a ia e analysis o mo ali y p edic o s was pe o med
92 using logis ic eg ession conside ing ha all 21 pa ien s we e ali e a day 30.
93
94 (b) conside ing he possibili y o up o 2 dea hs among hem ( wo- old he uppe limi o 99% CI). Dea h was
95 andomly assigned o 2 pa ien s om he 21 wi h 10 i e a ions, and mul i a ia e logis ic eg ession was
96 conduc ed o each i e a ion.
97
Page 99 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
5
98 Table S1. Fea u es o pa ien s acco ding o whe he ollow-up blood cul u es we e pe o med
99
100
FUBC pe o med
N=122
FUBC no
pe o med
N=323
OR (95% CI)
p- alue
Fe e >72h
27 (22.1)
50 (15.5)
1.55 (0.92 - 2.62)
0.098
Lack o ea ly ca he e
emo al (CRBSI only)
21/89 (23.6)
120/247 (48.6)
0.50 (0.33 - 0.77)
<0.001
Endoca di is
12 (9.8)
7 (2.2)
4.93 (1.89 - 12.82)
<0.001
Ca diac p os he ic al e
9 (7.4)
7 (2.2)
3.60 (1.31 – 9.88)
0.008
Pacemake
8 (6.6)
5 (1.5)
4.46 (1.43 – 13.92)
0.005
A icula p os he ic de ice
6 (4.9)
9 (2.8)
1.81 (0.63 – 5.18)
0.266
Endo ascula p os hesis
5 (4.1)
4 (1.2)
3.41 (0.90 – 12.91)
0.056
Me as a ic oci
3 (2.5)
4 (1.2)
2.01 (0.44 – 9.12)
0.356
SOFA sco e – median
(IQR)
3 (1 – 5)
3 (1 – 6)
0.97 (0.91 - 1.04)
0.407
101
Page 100 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60

Fo Pee Re iew
6
102
103 Table S2. Bi a ia e analysis o he associa ion o di e en exposu es wi h in-hospi al mo ali y.
104
Va iable
OR
95% CI
p- alue
Age, median (IQR)
1.03
1.01 – 1.05
<0.001
Age > 65
2.48
1.39 - 4.42
0.002
Male gende
1.26
0.74 – 2.15
0.402
Age-adjus ed Cha lson
1.07
0.90 – 1.26
0.442
Age-adjus ed Cha lson > 3
2.40
1.34 – 4.27
0.002
Diabe es melli us
1.03
0.50 – 2.14
0.936
Ch onic li e disease
0.82
0.31 – 2.17
0.683
Mode a e- o-se e e CKD
2.14
1.05 – 4.37
0.034
Hemodialysis
1.44
0.28 – 7.53
0.648
Ce eb o ascula disease
2.79
1.43 – 5.45
0.002
COPD
1.47
0.64 – 3.34
0.359
Ch onic hea ailu e
3.13
1.49 – 6.59
0.002
Rheuma ologic disease
0.51
0.07 – 3.98
1.000
Solid malignancy
1.27
0.73 – 2.29
0.416
Immunosupp essi e he apy
1.89
0.99 – 3.61
0.050
Recen an ibio ic he apy (30 days)
1.48
0.87 – 2.52
0.145
Recen su ge y (30 days)
0.53
0.27 – 1.03
0.049
Communi y-acqui ed
0.97
0.95 – 0.99
0.228
Hospi al-acqui ed
0.61
0.33 – 1.13
0.112
Heal hca e-associa ed
1.92
1.02 – 3.62
0.040
PICC
1.00
0.49 – 2.02
0.998
PVC
1.41
0.83 – 2.38
0.201
Sho - e m CVC
0.87
0.52 – 1.46
0.511
Tunneled ca he e
3.42
0.97 –
12.02
0.068
Po -a-Ca h
0.96
0.11 – 8.06
1.000
Pacemake
2.65
0.79 – 8.85
0.112
Vascula p os he ic de ice
1.66
0.34 – 8.15
0.628
A icula p os hesis
1.45
0.39 – 5.29
0.476
Ca diac al ula p os hesis
1.97
0.62 – 6.29
0.275
Pa en e al nu i ion
0.77
0.39 – 1.50
0.437
Sep ic shock
5.07
2.40 –
10.72
<0.001
SOFA sco e
1.20
1.07 – 1.34
0.002
SOFA sco e ≥ 2
1.92
1.11 – 3.35
0.019
Pi sco e
1.26
1.12 – 1.43
<0.001
Pi sco e > 3
3.57
1.57 – 8.12
0.004
ICU admission
1.75
0.97 – 3.15
0.072
Amine suppo
2.55
0.88 – 7.40
0.080
Oxacillin esis ance
1.54
0.73 - 3.27
0.260
Ca he e - ela ed ocus
0.55
0.31 – 0.96
0.034
Unknown ocus
1.14
0.51 – 2.55
0.755
U ina y ocus
0.99
0.98 – 0.99
0.348
Skin and so issues ocus
1.59
0.43 – 5.87
0.480
Endoca di is ocus
2.82
1.03 – 7.69
0.036
Respi a o y ocus
6.05
1.47 –
24.82
0.005
Abdominal ocus
0.96
0.11 – 8.08
0.967
Bone and join s ocus
1.15
0.13 –
10.01
0.899
No empi ic he apy ecei ed
1.03
0.59 - 1.79
0.922
Adequacy o empi ic he apy
0.79
0.42 – 1.50
0.468
Du a ion empi ic, days
0.99
0.97 – 1.01
0.481
Du a ion a ge ed, days
0.99
0.99 – 1.01
0.353
Page 101 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
7
Du a ion o he apy o e all, days
0.95
0.92 – 0.99
0.013
Ca he e emo al in CRBSI
0.59
0.31 - 1.11
0.101
Pe sis en e e a 72h
3.19
1.787 –
5.71
<0.001
Pe sis en bac e emia
2.74
1.350 –
5.57
0.005
Sep ic complica ions
4.45
0.974 –
20.37
0.036
Complica ed BSI
2.19
1.25 – 3.85
0.005
Days o ca he e emo al
0.99
0.84 – 1.17
0.888
ICU s ay, days
0.99
0.98 – 1.01
0.758
BSI onse o discha ge, days
0.99
0.99 – 1.01
0.327
Admission o BSI, days
1.00
0.99 – 1.01
0.447
105 CKD: ch onic kidney disease; COPD: ch onic obs uc i e pulmona y disease; AIDS: acqui ed
106 immunode iciency synd ome; PICC: pe iphe ally inse ed cen al ca he e ; PVC: pe iphe al enous ca he e ;
107 CVC: cen al enous ca he e ; ICU: in ensi e ca e uni ; CRBSI: ca he e - ela ed bloods eam in ec ion; BSI:
108 bloods eam in ec ion
109
Page 102 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
8
110 Table S3. Mul i a ia e analysis o ac o s associa ed wi h in-hospi al mo ali y.
111
112 Model 1
113
Va iable
OR
95% CI
p- alue
Age > 65 yea s old
2.87
1.53 – 5.38
< 0.001
Ce eb o ascula disease
3.55
1.72 – 7.35
< 0.001
Immunosupp essi e he apy
3.24
1.55 – 6.79
0.002
Pi sco e > 3
3.54
1.45 – 8.64
0.005
Complica ed BSI
1.99
1.10 – 3.59
0.023
114 BSI: bloods eam in ec ion
115 AUROC o 0.74 (95%CI 0.67 – 0.80)
116
117 Model 2
118
Va iable
OR
95% CI
p- alue
Age > 65 yea s old
2.99
1.57 – 5.71
< 0.001
Ce eb o ascula disease
3.59
1.71 – 7.53
< 0.001
Immunosupp essi e he apy
2.98
1.34 - 6.41
0.005
Pi sco e > 3
3.52
1.38 – 8.97
0.009
Fe e > 72 hou s
3.29
1.72 – 6.28
< 0.001
Ea ly ca he e emo al in CRBSI
0.50
0.28 - 0.90
0.020
119 CRBSI: ca he e - ela ed bloods eam in ec ion
120 AUROC o 0.76 (95%CI 0.69 – 0.82)
121
Page 103 o 108 Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Fo Pee Re iew
9
122 Table S4. Sensi i i y analysis o 30-day mo ali y p edic ion.
123
124
125 (a) Conside ing ha all 21 pa ien s su i ed a day 30
126
Va iable
OR
95% CI
p- alue
Age > 65 yea s
2.47
1.39 – 4.38
0.002
Ce eb o ascula disease
3.07
1.53 – 6.19
0.002
Immunosupp essi e he apy
2.92
1.46 – 5.85
0.003
Pi sco e > 3
3.39
1.43 – 8.00
0.005
Complica ed BSI
1.95
1.12 – 3.37
0.017
127 BSI: bloods eam in ec ion
128 AUROC 0.72 (95%CI 0.66 – 0.78)
129
130 (b) Conside ing ha up o 2 pa ien s o hem died (chosen andomly). We p o ide a summa y o
131 he es ima es in he 10 models.
132 1.
Va iable
OR
95% CI
p- alue
Age > 65 yea s
2.47
1.40 – 4.38
0.002
Ce eb o ascula disease
3.07
1.53 – 6.19
0.002
Immunosupp essi e he apy
2.92
1.46 – 5.85
0.003
Pi sco e > 3
3.39
1.43 – 8.00
0.005
Complica ed BSI
1.95
1.12 – 3.37
0.017
133 BSI: bloods eam in ec ion
134 AUROC 0.73 (95%CI 0.66 – 0.78)
135
136 2.
Va iable
OR
95% CI
p- alue
Age > 65 yea s
2.58
1.46 – 4.57
0.001
Ce eb o ascula disease
2.93
1.46 – 5.91
0.003
Immunosupp essi e he apy
2.87
1.43 – 5.75
0.003
Pi sco e > 3
3.90
1.67 – 9.12
0.002
Complica ed BSI
1.90
1.10 – 3.25
0.021
137 BSI: bloods eam in ec ion
138 AUROC 0.73 (95%CI 0.66 – 0.78)
139 3.
140
Va iable
OR
95% CI
p- alue
Age > 65 yea s
2.59
1.47 – 4.59
0.001
Ce eb o ascula disease
3.33
1.67 – 6.67
< 0.001
Immunosupp essi e he apy
2.84
1.42 – 5.69
0.003
Pi sco e > 3
3.22
1.36 – 7.63
0.009
Complica ed BSI
2.01
1.20 – 3.59
0.009
141 BSI: bloods eam in ec ion
142 AUROC 0.73 (95%CI 0.67 – 0.79)
143
144 4.
145
Va iable
OR
95% CI
p- alue
Page 104 o 108Clinical Mic obiology and In ec ion
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Mortality predictors and definition proposal for complicated coagulase-negative Staphylococcus bacteraemia: a multicentre prospective cohort study

Abstract

Mortality predictors and definition proposal for complicated coagulase-negative Staphylococcus bacteraemia: a multicentre prospective cohort study
