Content uploaded by Fortino Solórzano-Santos
Author content
All content in this area was uploaded by Fortino Solórzano-Santos on May 02, 2018
Content may be subject to copyright.
r.i,
E'i
Ei
H{
Ix
E{
EI
Ffi
r.{
i-t
t-2
f .: .4.
icl',1
rli
E}
B"- i
!$
::;
rlt
;il
v|^
A,jI
i. ,I
1..,1
,.,t
it
}j.,
F..
8..
Í:,
ir,
r-;,
l.' '
i.l
i:,
i.
ii
t
iit
¡n
--.--.n-:. /:-r,* l
i obs2 ¡,Sc-;rp / s I rc¡1 ur.eek- - 4.1
it:
,,-J
S;¿nd J inre;i Dis l0 00-00. ]a9i ORIGINAL ARTICLE
A Blood Micro-Culture Systern for the Diagns§ls
of Bacterelr:ia Ped!atric Patients
FORTINO SOLORZANO.SANTOS,, IVIARIA GUADALUPE I\{iRANDA-NOVALE'S],
BLANCA LEAJ'{CS.}¡1IRANDA2. HUI\{BERTC DIAT-PO}iCtrI ATId
GERARDO PALACiOS-SAUCEDO 1
Front thetlr¡fectious Diseases Departntertr. and the 2illrcrobiologt Laborctoríes. Pediatr¡c Hosoial National ]ledical
Ceutr, 'Sigli XXl. Social Serl,rit.v i'{exiccn Insrirute, folexíco Cit's }'fexico
The aim of this study Eas to eysluate tbe utilit-v of a rolume-modified blood culture svlem to diagnose bacteremia in
neryborns:¡nd infents. A total of?93 paired blood cultures. obt:rined from {6-{ patients (173 nenborns and 291 infants)' rvere
analyzed. YacuteinerD tubes containing 18 ml supplement«i peptone broth sodium-polyanethoi-sulfonate were used as the gold
standard. in comparison with a blood micro-culture s¡'sten, containing l'8 m! of the bloth' Prior to ¿ntibiotic treatment' 2'2
ml of blood ¡,as obtained from each patient; 0.2 ml rvas inoculated in a biood micro-culture tube and 2 ml in a routine tube'
Sensitirirr,. specificity and predictire "alues *ere calculated. Illicroorganisms lrere isolated in 153 st¡ndard blood'cuiture tubes
anrl 151 blood micro-culture tubes. The scnsitivis of the Lrlood micro-culture svstem was 95%. specificitl' 99% and positive
andnegativepredictivevaluesg6T" a*dggoírespeciively.Thesensitivir¡'andspecificit-vofbloodmicro-cultureinneonatesand
infanls is high. \ve rerommend. th3t this system be usei for rhe dirgsosis of bsctcremi: io nc\illorüs and infants in laboratori¿s
where manuai slstems are still in use.
F. Solor:ono santos, fufD, Departmento dc Infecrologia,Hospitat de Petliatía. c¿ntro Médico l¡'acional sxxl' Instituto
ilfexitano del Seguro Socíat. )r. Cuauhtentoc 330, Col. Docrores, CP 06720. l'Iexico DF' Mexico
tems) were used as ihe gold standard They are used as the routine
blood culture in this Peciiarric Hospital The blood micro-cultu¡e
system contained 1.8 rnj oi ihe same supplemented peprone brorh
in 12 x ?5-¡nm s',e.ile \'¿cuum tubes (Becton Dickinson' Diagrostic
S;tster,rs). A1l blooci i-nicro-cuirure tubes were incubateci for 96 h at
35"C and inspecteci before use ro exclurie contamtnatlon
Beio¡e t:'eal¡nent with anlibiolics, 1-3 blood samoles rve¡e
drawn lrom each paiieili. Approximately ?'2 ml blood rvas ob'
tained by percuianeous venipuncrure after skin preparation with
l0ozi povicione-iodine lollowed by TAoiL isopropyi alcohol' A vol-
,-. of 0.2 ml pas inoculated inro a blood micro-culture tube and
tire remaining 2 ml into a routine blood culture tube' These culiu¡e
sampies. obáinei iron the same d¡arv' were conside¡ed paircci'
Culiures.noi paired were discarded. Al1 cuhu¡es we¡e Droc¿sseri
according- to the sta¡¡iar<ls set oul by the American Society ior
It{icrobioiogy (9).
A1l tubes were vented when they a¡rived at rhe laboraiory and
incubated ai i5"C. Tubes showing macroscopic evidence olgTo*lh
(rurbidity or hemoi,vsis) on daiiy inspection, were Gram-stained
and subcultured aero'oicaiiy on a sheep blood agar plate' a Mac-
Conkey agai plate and 2 chocolaie agar piates' One chocoiate agar
piu,* iurln"ubr¡ed in a microaerophiiic atmosphe:e and the other
in an anaerobic aimosphere at i5"C. Culiu¡es '¡'ith no macroscopic
evicience of growlh were subcultured blind on days I' 2 anC 7 ani
',rere Crscarced if ..he last subcultu¡e uas nee:tire'
A neonarologist oi pediatrician and a pediatric infecticus disease
specialist ju<iged rhe ilinical sigoificance of the mic¡obial isolates'
p,r¡tirn.A- g,rid.lin*, -used to consider an isolaie as signiñcant
include the- lollowing criteria: predisposing lactors' ciinical signs
and symptoms; abnorrnal compiete biood countl iadiographic ñn<i-
i"*,-'rá0., ol positive biáod cultures; species ident-ifrcation;
,sáation of the szLme species from anoiher body site; and response
to antimicrobiai theraPY.
A blood mic¡o-cuiture result was considereC a laise posilive if
bacterial growth was present in the micro-culture' but not in tbe
gold sranáard. Any iate growth (by dzy 7) n bir¡od micro-cul:ure
was aiso considered a false positive resuir' A result was consiri¿red
faise negative when the bláod mic¡o-cultu¡e iailed to isoiate the
g)
TNTRODUCTION
Biood cuiture is one of the most important proceciures
performed in the clinjcal microbiologl' laboratory- Detec-
.,ion oi bacte¡emia or iun-qemia guicies the clinician ln
choosing the best antimic¡obial treatment. Bacte¡emia ca¡
be continuous o¡ intermittent and the sequence is dilficult
to predict, thus it is necessary io take more ihan 1 sample
for bloori culruring (1. 2).
Ferv clinicians are aware of the porverfui inffuence that
the volume ol blood has on the sensitivity ol blood cuitu¡e'
The best ./olums needed fo¡ blcod cuiture has no¡ been
est¿Lblisheci. A volume of 10-20 ml pei sample has been
recomrnended fo¡ adult patients and 30 n-rl for lebrile
immunocompromised patients and thcse rlith endocarclitis
(1, 3). Aithou.-eh 1-5 m] is consideied an optimal volun-re
tor children (1,5), it is possible that such.low volumes oi
blood mi-eht reduce lhe sensitivity of rhe blood culiures
(6-8).
Because in some peciiatric patieots. especiali,v newborns
and iniants, it is difficult to obtain the required amount oi
biood per drarv lo¡ the biood cuiture s.vslems currenili'
available, we evaluated the utiiity ol a volum¿-modif,erl
blood culture system requiring a blood sample volume of
only 0.2 m1 lor ciiagnosis of bacieremia.
IvÍATERIALS AND IVÍETHODS
Iafants l¡om birth to 12 months of age *'irh ciinicai signs and
slmptoms of sepsis were included in the study- Verbal consent was
ob',ained l¡om the parents of each patient. The study was approved
by the local Ethics and Resea¡ch Committee.
Vacutaine¡o tubes containing l8 nl supplemented peptone broth
sodium-polyanethol-sulionate (Becton-Dickinson. Diagnostic Sys-
O 1998 Scandinavian University Fress. ISSN 0036-5543
7/ !rbL: iilto :,
2 F. Solarzano-Santos et al. Scand J infxt Dis -10
\--l Microorsanism Standard blood
culture
Table I. Comparison o7'microorgantst¡ts isoiated in 108 newborns
and infants using 2 blood cuhure sy'stems erowths were false positive (3 Bacilius sp. and 2 CNS late
_qrowlhs, i S. aureus an{i 1 CNS earlv erowths) (Tabie i).
The number oi cui¡ures try age group is shorvn in Tabie
II. The sensitivity of the blood micro-culture system was
959'., with a specificity of 99%, and a posirive predictive
value (PP\) and a negative predictive value il.lPV) ol 96Y,
and 99Yo. respectively. For newborns, sensitivity and spe-
cificity w'ere 96'% and 99%, with a PPV of 980/o and NPV of
99%. For infants, sensitivity and specif,city were 94oA and
99ozi,. with PPV oi 96oA and NPV of 98%.
The microorganisms isolated were Gram-positives 510A,
Gram-negatives 42oA, anaerobes 3oA and yeasts 3% (Table
II). Both culture systems took 24-48 h to show growth.
Tlere rvas no signiñcant diflerence between the 2 s¡,stems in
this respect. I¡ 920A it took 24 h to show grorvth and in 89'o
48 h. ñ{icroorganisms that grew at 48 h in both systems
rvere H. influenzae (1), B.fragilis (1), Bacteroides sp. (2) and
Candida sp. (3) Blood micro-culture faiied to isolate 2
enterobacteria (1 E. coli and 1 S. enteritidis), 3 S. aureus, 1
CNS and 1 E. faecalis.
DISCUSSION
Diagnosis of bacteremia depends on the demonstration of
the presence ol microorganisms in the blood. The drawing
of 2 or 3 samples from different venipunctures is currently
recommended ior blood culturing, in order to increase the
probability of detecting intermittent bacte¡emias or fun-
gemias and the probability of correct diagnosis of bac-
teremia (10, 1l). Because of the iow number ol
colony-forming units per mI blood (10, 12), the yield of
blood cultures is clearly volume-dependent even in adults.
In infants and child¡en. the number of coiony-forming units
per ml biood tends to be far greater than in adults, thus
smaller volumes (1-3 nü) of blood appear to be sufficient
for blood culture. Nevertheless, the optimal volume and
number of sampies for blood culture in children is still
controversiai (3, 6, 8, 11).
Pediatric biood culture systems currently in use require
betrveen 0.5 ml and 2 ml of blood per sample. Drawing
such a biood volume is difficult in many pediatric parients
because of thei¡ 1ow intravascula¡ volume, especíaliy in very
low birthweight newboms. Thus, many pediatricians obtain
only a single blood sample, usually of I ml or iess (13).
Because ol the high yieid of our blood micro-culture sys-
tem, we propose to use it in very smali and severely ill
Blood-lmicro-
cuiture
Gram-positives
CNS*
S. aureus
S. pneumoniae
E. laecalis
Bacillus sp.
S. viridans
S. agalactiae
Gram-negatives
E. coli
S. enteriridis
K. pneumoniae
Non lermentative
GNBb
Pseudomonas spp.
E- aerogenes
K. oxytoca
H. influenzae
Anaerobes
B. fragilis
Bacteroides sp.
Yeast
Candida aibicans
Candida sp.
25
Z7
8
9
4
3
2
T,1
15
l4
6
5
5
I
1
1
27
25
8
I
(
,
i6
14
14
6
5
5
1
1
1
3
2
3
2
3
2
CNS : coagulase negative Staphylococcus.
CNB : Gram-negative bacilli.
bacteria isoiated in its paired gold standard tube. For statistical
analysis, sensitivity, specificity and preciictive vaiues of the biood
mic¡o-culture system were caiculated.
RESULTS
A total ol 793 paired blood cultures obtained l¡om 464
patients (173 newborns and 291 infants) were anaiy'zed.
Pathogens were recovered f¡om 108 patients (39 newboms
and 69 infants). lv{icroorganisms were isolated in 153 stan-
ciard blooci cuiture tubes and 151 blood micro-culture
tubes. Growth in 7 blood mic¡o-cultures were false nesative
(coagulase negative Staphylococcus (CNS) 1, Staphylococ-
cus aureus 3, Enterocorcus faecalis i, E. coii 1 and S.
enteritidis 1). On the other hand, 5 blood mic¡o-culture
Table II. Standard us. blood micro-cultures according to age group
)
2
Standa¡d bloori cultu¡es Biood mic¡o-cui*.ure Totai paired biood cultu¡es
e--\
,Ga*ñ-- \
--d.ge group (x) Positive Ne.-eative Positive Negative !--.-
.J - _l
Newborn (l?3)
iniaats (291)
Total (464)
56
97
t53
232.
408
640
55
96
1<l
288
505
793
233
409
&2
' /psc:--ct5 : ¡ scc4r¡olsl ;S.ll?,rS t::r'.üc€k. :+ Ps-:-i137?. CC: iicr No-¡
Scand J Inf*t Dr l0 Blood-mícrocttiture sr-sten for bacteremia 3
infants. The smaller blood volume per viai (0.2 nI) .equired
for the blood micro-cuiture. allows the clinician to draw 2
o¡ rirore samples in these patien¡s, avoiciing the dsk ol
depleting their intravascular volume. in spite ol the high
sensitivity and specif,city we lound for blooci micro-culture
in neonates and infants, taking 2 or 3 blood samples fi-om
a patient couid increase bacterial detection, which would
help the clinician to choose the correct antibioiic t¡eátment.
Dillerent studies have shown that a blood-to-broth iatio of
aI leasi 1:10 is optimal lor bacierial detection (14). Our
resuits show that if we maintain this ratio, the volume ol
biood required for each cuiture tube can be low, a condi-
tion that other authors have addressed previousiy (13, 15)-
The sensitivity and specificíty of the blood micro-culture
system for anaerobes and yeasts were simiiar to ¡outine
biood culture results, which implies the possibiiity oi using
this system even in cases in '"vhich ihese etioiogic a_qents are
suspected.
.A,rtom?.tic blood culture systems are nct avaiiable in
many iaboratories in Latin America, mainly because of the
cost of this equipment, and even when the automated
detection sy-stems are donated by the manufacturer the
institutions cannot always alford to purchase all the botties
reqr-rired. For these laboratories, this micro-culture method
could be a good alternative. We recommend it fo¡ the
diagnosis of bacteremia in rrewborcs and infants, fo¡ use in
laboratories whe¡e manuai systerns are still being used.
Automatic blood cuiture systems currently in use are not
compaiible rvith this blood micro-cul¡ure method; a further
invesiigarion will be required to evaiuate different blooci-to-
broth ratios in commercial blood cullure sysiems.
REFEREhICES
1. Washinglon II JA. Blood cultures. Principles anC techniques.
iv4ayo Ciin Proc 1975; 50: 9i-7.
5.
Brc*.n WR. Fxtrapolating to bacterial life outside the test
tubr. J Antinic¡ob Chemother l99l; 27: 565 1.
Shanson DC. tslood cuiture iechnique: cuüerrt coniroversies.
J Antimicrob Che¡¡other 1990: 25: 1?-29.
Fox H. Evaiuation of use of sigaal system blood cuitures in
paediatrics. J Clin Pathoi i988: 4l: 683-6.
Szymczak EG. Barr WA, Coldman DA. Evaluation of
blooC cuiture prccedures in a pediatric hospital. J Ciin lv{i-
crobiol 1979: 9: 88-90.
ó. Tenne¡, JH, Reller LB. Mirret S. Wang W, Weinstein MP.
Controlled evaluation of the volume ol blood cuitured in
detection oi bacteremia and luneemia. J Clin Microbiol
1932: l5: 558-60.
11.
9.
8.
Mermei LA, l'faki DG. Detection of bacteremia in adulrs:
consequences of culturing an inadequate volume of blood.
Ann Intern }fed i993; Ll9 270-2.
Kellog JA. Ferrentino FL, Goodstein MH, Liss J. Shapiro
SL, Bankert DA. Frequency oi low level bacteremia in in-
fants from birrh to two months of age. Pediatr Infect Dis J
1997;16:381-5.
Reller B. Murray P, Mac Lowry J. Blood cuitures II. Curru-
lative techniques and o¡ocedures in clinical microbioiogy.
Cumitech 1 A. American Society for Microbiology. 1982:
l tt
Washington JA, Ilstrup DM^ Blood cultures: Issues and con-
troversies. Rev Infect Dis 1986; 8:792-802.
Isaacman DJ. Karasin RB, Reynolds EA, Kost SI. Eflect of
number of blood cultures and volume of blood on detection
of bacte¡emia in chiidren. J Pediatr 1996; 128: 190-5.
Ilstrup DM, Washington lA. The importance of volume ol
blood cul¡ure in the detection ol bacteremia and fungemia.
Diag lv{icrobioj Inlect Dis l93i; i: i07-10.
Neal PR, Kleiman M, Reynolds IK, Allen SD, Lemons JA.
Yu PL. Voiume of blood submitted for culture lrom
neonates. I Clin Microbiol 1986: 21: 353-6.
Auckenthale¡ R; Ilstrup DNf, \l'ashinglon JA. Comparison
of recovery oi organism from blood culture diiuted (volumei
volume) ana 2OoA (volume/voiume). J Clin Microbiol 1982;
l5:860-4.
15. Mangurten HH, LeBeau LJ. Diagnosis of neonatal bac-
teremia by a microblood cullure technique. J Pediatr 1977;
90: 990-2.
Submitted Julv 13, 1998 accepted Actober 9, 1998
10.
12.
t3
14.
