1
Evaluation of the Blood Volume Effect on the Diagnosis of Bacteremia in
2
Automated Blood Culture Systems
3
Hsiu-Hsien Lina, Yen-Fang Liub, Ni Tiena, Cheng-Mao Hoa,c, Ling-Nu Hsub,
4
Jang-Jih Lua,c,d*
5
6
a
7
Taiwan
8
b
9
c
Department of Laboratory Medicine, China Medical University Hospital, Taichung,
Department of Nursing, China Medical University Hospital, Taichung, Taiwan;
Graduate Institute of Clinical Medical Science, China Medical University, Taichung,
10
Taiwan
11
d
12
Taoyuan, Taiwan
13
Running Title: Blood Volume Effect on Bacteremia Diagnosis
14
*Corresponding author:
15
Dr. Jang-Jih Lu, Department of Laboratory Medicine, China Medical University
16
Hospital, No.2, Yuh-Der Road, Taichung, Taiwan 40447.
17
E-mail: janglu45@gmail.com
Department of Laboratory Medicine, Linkou Chang-Gung Memorial Hospital,
1
1
Abstract
2
Background: Blood culture volume is the most important variable in detecting
3
bacteremia and fungemia. However, the majority of hospitals in Taiwan do not meet
4
the criteria of ideal blood culture volume (8-10ml per bottle, two bottles per set)
5
during collection.
6
Methods: The object of this study is to initiate an educational program for health-care
7
workers to increase blood volume collection and to evaluate the relationship between
8
blood volumes and recovery rate for detecting bacteremia and fungemia effectively by
9
using BD BACTEC 9240 blood culture system.
10
Results: After education, the≧ 5mL group increased from 2.93% to 71.24%. For total
11
4,844 bottles, the relative improvement for recovery rate has increased 17.81%
12
between<5 mL group and≧5 mL group. The recovery rates of low volume (<3 mL),
13
mid volume (3-7 mL), high volume (8-10 mL) and extreme high volume(>10 mL)
14
groups are 13.31%, 15.02%, 17.68%, and 14.96%.
15
Conclusions: With good blood collection practice, our study found blood volume has
16
direct proportion to recovery rate.
17
18
KEYWORDS
19
Automated blood culture; Blood culture volume; Recovery rate
2
1
Introduction
2
Among many kinds of infection, blood-stream infection is systemic and could result
3
in life-threatening sepsis, which lead to high morbidity and mortality, therefore, early
4
and accurate detection is important to improve patient care1. Positive blood culture
5
could help to diagnose bacteremia and fungemia among patient with fever, because
6
the number of microorganisms present in the adult’s blood is typically low-level,
7
usually fewer than 10 CFU/mL, sometimes less than 1 CFU/mL. The volume of blood
8
for culture is therefore the most important variable in determining the recovery rate of
9
blood culture2. Due to the collection device limitation, despite the importance, many
10
hospitals in Taiwan could not meet the recommended volume, which should be 20-30
11
mL per venipuncture3. Our preliminary study shows that the average blood volume for
12
blood culture in China Medical University Hospital is low, less than 3 ml per bottle in
13
89.1% of venepunctures. The major reason would be that blood culture usually comes
14
with multiple tubes collection; therefore, one-10 ml syringe can only deliver about 2-3
15
mL blood into blood culture bottles. Other reasons are like the traditional concept of
16
Taiwan people, and most of Asian people think that the blood is too precious to be
17
drawn. The current method for blood culture is automatic detection, which should
18
more sensitive than traditional manual method4.
19
In this study, we evaluated the importance between blood volume and recovery,
3
1
and the relative detection time in automatic blood culture system. Our design for this
2
study is to institute a series of educational programs for health-care workers,
3
regarding to adequate blood collection practice. Afterward we calculated the blood
4
volume and organism distribution, and then compared the recovery rate and
5
microorganism species between low, mid and high volume group. We wanted to
6
determine if there is clinical significance and thereby convince the patients who need
7
blood cultures that sufficient volume of blood is of clinical importance.
8
9
10
11
12
13
14
15
16
17
18
19
4
1
Methods
2
Educational program. During July 2010, several courses were held in the ER unit
3
focusing on adequate blood collection, which included disinfection, needle stick
4
prevention by using safety blood collection sets (BD Safety-Lok™), which can allow
5
serial blood collection and let each tube or bottle reach ideal volume. Blood volume
6
inoculation practice into blood culture media sets were performed according to
7
manufacturer’s instructions (BD BACTEC™ Standard aerobic/F medium, BD
8
BACTEC™ lytic/Anaerobic/F medium).
9
Collection of samples. For a 6-months period, a total of 4,844 samples were collected
10
in the ER unit, the volumes of blood collected were measured by performing a
11
comparison with reference bottles which had been injected with 1 to 10 mL colored
12
solution in advance. All of the bottles were placed into the BD BACTEC™ 9240
13
instruments (BD Diagnostic Systems, Sparks, Md.).
14
Detection of samples. The BD BACTEC™ 9240 blood culture instruments are
15
designed for the rapid detection of microorganisms in clinical blood specimens. The
16
sample to be tested is inoculated into the vial which is entered into the BD
17
BACTEC™ instrument for incubation and periodic reading. Each vial contains a
18
sensor which responds to the concentration of CO2 produced by the metabolism of
19
microorganisms. The sensor is monitored by the instrument every ten minutes for an
5
1
increase in its fluorescence, which is proportional to the increasing amount of CO2
2
present in the vial. A positive reading indicates the presumptive presence of viable
3
microorganisms in the vial. The protocol length was a total of 5-days before a vial was
4
determined to be negative.
5
Identification of samples. All the positive vials were stained and identified by BD
6
Phoenix™ Automated Microbiology System (BD Diagnostic Systems, Sparks, Md.).
7
The BD Phoenix™ Automated Microbiology System is intended for the in vitro rapid
8
identification (ID) and quantitative determination of antimicrobial susceptibility by
9
minimal inhibitory concentration (MIC) of clinical frequently isolated
10
microorganisms.
11
12
Analysis of data. The paired t-test was used to verify effects on educational programs.
13
Relative improved recovery (R.I.R) was used to analyze relative recovery rate
14
between <5ml group and ≧5ml group under different circumstances. Chi-square and
15
ANOVA analysis were used to compare separate pair group and overall recovery
16
difference from low (<3 mL), mid (3-7 mL), high (8-10 mL), and extreme high (>
17
10mL) volume groups.
18
19
6
1
Results
2
Of the 4,844 samples submitted during the study period, we first compared blood
3
volume distribution before and after education. It was found that the≧5 mL group has
4
increased from 2.93% to 71.24%; whereas the <5 mL group has decreased from
5
97.07% to 28.76% (Fig. 1.), which shows a statistically significant difference (p<0.05).
6
Next, to know the recovery rate in the two groups with more precise criteria, we
7
excluded either duplication or contamination samples, or both, to see the differences
8
in relative improved recovery (R.I.R). From the data we found that R.I.R for total
9
samples, samples exclude duplication, samples exclude contamination, samples
10
exclude both duplication and contamination are 31.08%, 36.37%, 12.91% and 17.81%,
11
respectively (Table 1).
12
We also checked organism distribution between <5ml group and ≧5ml group,
13
the top two organisms for both groups are Enterobacteriaceae spp. and
14
Staphylococcus aureus (Table 2). The average time to detection for <5 mL group and
15
≧5 mL group are 17.9 and 17.6 hours (Table 3). To further analyze volume effect on
16
blood culture, we divided all the samples into low volume (<3 mL), mid volume (3-7
17
mL), high volume (8-10 mL) and extreme high volume group (>10 mL); the recovery
18
rate for those four groups above are: 13.31%, 15.02%, 17.68%, 14.96% (Fig. 2), by
19
pairing comparison, we found there is significant difference between low and high
7
1
group (p<0.05); by overall comparison, it also show significant difference among all
2
four groups (p<0.05).
3
4
8
1
Discussion
2
Obtaining the adequate blood volume plays an important role in the detection of
3
bloodstream infections5. We continue to educate our associates with this concept.
4
Through the educational program, blood volume collected during this period showed
5
significant increases in the ER unit, which demonstrated that our program is effective;
6
continuous training will be executed since adequate volume collection in Asian
7
countries is not easy due to traditional concept. Our analytic data also proved that
8
blood volume range from 8-10 mL per bottle demonstrated the best recovery rate,
9
comparing to 3-7 mL group and <3 mL group. We may use this data to convince our
10
patients to give higher volume of blood. We found that blood volume measurement by
11
visual comparison maybe a little subjective; for there might be deviation judging by
12
naked eyes. Weighting the bottle before and after sample collection would be the more
13
precise way to do6. According to previous study, more significant yield could be
14
reached if the low volume group (<3 mL) sample number could be lower, by which
15
the yield of blood cultures in adults increases approximately 3% per milliliter of blood
16
collected7.
17
In Table 1, the recovery difference between < 5 mL and ≧5 mL group seems to
18
be less significant when taking out contamination factors, which means our
19
contamination rate was high. One reason to such finding may due to syringe usage.
9
1
Nurses will change needle while injecting blood into different bottles, this practice not
2
only increase the chance of a needle stick injury, but also increase the chance of a
3
contaminate microorganism carried by gloves. To minimize man-made contamination,
4
we encourage our associate to apply blood collection set to blood drawing process,
5
with no need to change needle. When it comes to multiple tubes blood collection
6
combines with blood culture collection, we also encourage the usage of safety
7
collection device, which not only prevent needle stick injury and ensure adequate
8
volume was collected8, but also ease patient’s tension about large-volume syringe
9
usage. However, those isolates may be true pathogens because we did not verify if
10
11
there is only one bottle of a series of blood culture specimen identified9.
As for the recovery, we found that bottles that had been inoculated with higher
12
volumes of blood had higher recovery rate, which respond to previous
13
publications5,6,10,11 . From previous studies we know that blood culture set number is
14
also important to recovery; we may include this factor in our future study. Increasing
15
the number of blood culture sets drawn will also increase the volume of blood
16
collected, which should result in a further increase in positivity rates. Further analysis
17
to verify correlation with patient severity and recovery rate could be done for there are
18
studies addressed that severe patients had positive correlation between blood volume
19
and recovery rate; while mild patients did not6.
10
1
We found there is no significant difference in average detection time between <5
2
mL and ≧5 mL group, however, by organism sector, detection time of
3
Enterobacteriaceae genus in ≧5 mL group was faster than <5 mL group. Because
4
Enterobacteriaceae are the most frequent isolated microorganisms in blood culture, we
5
think it’s with clinical importance that higher volume contributes to faster time to
6
result. Our data in Table 3 indicated that for Enterobacteriaceae spp, <5 mL group
7
TTD was 11 hr while ≧5 mL TTD was 7.2 hr, with clinical significance. In our
8
hospital, 2010 Top 5 organisms in blood culture are CoNS, E.coli, K. pneumoniae,
9
Sta.aureus and MRSA. In conclusion, increasing blood volume has benefit both on
10
11
recovery and early detection.
There is one interesting finding in our study, for the extreme high (>10 mL)
12
volume group, the recovery rate was decreased compared to high (8-10ml) volume
13
group, the possible cause may be the dilution effect, which means failure to maintain
14
blood to broth ratio 1:5 to 1:10 may result in false negative result12, because of the
15
complement, lysozyme, phagocytes, antibodies, and antimicrobial agents inhibitory
16
effect13,14. This problem can be solved by using neutralizing agents such as
17
resin-containing culture media, which inactivate the inhibitory substances present in
18
blood.
19
Development of automated continuous-monitoring blood culture system during
11
1
the 1990s accelerated the trend from conventional manual methods. However, the
2
fundamental variables those provide accurate data for blood culture method remains
3
important.
4
In conclusion, adequate blood volume is correlated with increased recovery rates
5
of blood culture. Therefore, we must continue educating the phlebotomist about the
6
concept of collecting adequate blood volume to improve the recovery rate. In this way,
7
the infections of patients with sepsis can be detected rapidly and correctly. This will
8
benefit both patients and hospitals.
9
12
1
2
Acknowledgments
This work was supported by grants from the National Science Council (NSC
3
98-2320-B-039-032-MY3), China Medical University Hospital (DMR-100-128). We
4
thank Dr. Michael L. Towns for assistance with the manuscript.
5
6
13
1
Reference
2
1.
3
4
Reimer LG, Wilson ML, Weinstein MP. Update on detection of bacteremia and
fungemia. Clin Microbiol Rev 1997;10:444-65.
2.
Arpi M, Bentzon MW, Jensen J, Frederiksen W. Importance of blood volume
5
cultured in the detection of bacteremia. Eur J Clin Microbiol Infect Dis
6
1989;8:838-42.
7
3.
8
9
Infect Dis 1986;8:792-802.
4.
10
11
Weinstein MP. Current blood culture methods and systems: clinical concepts,
technology, and interpretation of results. Clin Infect Dis 1996;23:40-6.
5.
12
13
Washington JA, 2nd, Ilstrup DM. Blood cultures: issues and controversies. Rev
Li J, Plorde JJ, Carlson LG. Effects of volume and periodicity on blood cultures.
J Clin Microbiol 1994;32:2829-31.
6.
Ilstrup DM, Washington JA, 2nd. The importance of volume of blood cultured in
14
the detection of bacteremia and fungemia. Diagn Microbiol Infect Dis
15
1983;1:107-10.
16
7.
17
18
19
Mermel LA, Maki DG. Detection of bacteremia in adults: consequences of
culturing an inadequate volume of blood. Ann Intern Med 1993;119:270-2.
8.
Mendelson MH, Lin-Chen BY, Solomon R, Bailey E, Kogan G, Goldbold J.
Evaluation of a safety resheathable winged steel needle for prevention of
14
1
percutaneous injuries associated with intravascular-access procedures among
2
healthcare workers. Infect Control Hosp Epidemiol 2003;24:105-12.
3
9.
Bekeris LG, Tworek JA, Walsh MK, Valenstein PN. Trends in blood culture
4
contamination: a College of American Pathologists Q-Tracks study of 356
5
institutions. Arch Pathol Lab Med 2005;129:1222-5.
6
10. Cockerill FR, 3rd, Wilson JW, Vetter EA, Goodman KM, Torgerson CA,
7
Harmsen WS, et al. Optimal testing parameters for blood cultures. Clin Infect
8
Dis 2004;38:1724-30.
9
10
11
11. Hall MM, Ilstrup DM, Washington JA, 2nd. Effect of volume of blood cultured
on detection of bacteremia. J Clin Microbiol 1976;3:643-5.
12. Bouza E, Sousa D, Rodriguez-Creixems M, Lechuz JG, Munoz P. Is the volume
12
of blood cultured still a significant factor in the diagnosis of bloodstream
13
infections? J Clin Microbiol 2007;45:2765-9.
14
13. Tenney JH, Reller LB, Mirrett S, Wang WL, Weinstein MP. Controlled
15
evaluation of the volume of blood cultured in detection of bacteremia and
16
fungemia. J Clin Microbiol 1982;15:558-61.
17
14. Auckenthaler R, Ilstrup DM, Washington JA, 2nd. Comparison of recovery of
18
organisms from blood cultures diluted 10% (volume/volume) and 20%
19
(volume/volume). J Clin Microbiol 1982;15:860-4.
15
1
2
Case no.
2500
*(P<0.05)
2000
3
4
1500
1000
500
5
0
6
Before Education
7
Figure 1. Blood Volume Analysis.
8
Comparison was made by < 5mL(■) group and ≧5 mL(■) group. After education, the
9
≧5mL group had increased from 2.93% to 71.24% and the <5ml group has decreased
10
After Education
from 97.07% to 28.76%.
11
12
16
1
2
3
4
5
6
7
* (p<0.05)
20%
18%
16%
14%
12%
10%
8%
6%
4%
2%
0%
Low vol.
Mid vol.
(<3 ml)
(3-7 ml)
High vol.
(8-10 ml)
Extreme vol.
(>10 ml)
8
Figure 2. Recovery rate analysis. Comparison was made between low volume (<3
9
ml), mid volume (3-7 ml), high volume (8-10 ml) and extreme high volume(>10 ml)
10
groups to see the recovery rate.
11
12
13
14
17
1
Table 1. Relative improved recovery rate analysis between <5ml groups and
2
≧5ml groups.
Volume
Recovery rate
Total
Exclude dup
Exclude conta
Exclude conta & dup
<5ml
13.20%
14.60%
11.39%
12.07%
≧5ml
17.30%
19.91%
12.86%
14.22%
R.I.R
31.08%
36.37%
12.91%
17.81%
3
R.I.R: Relative improved recovery; dup: Duplicate isolates; conta: Contaminated
4
bacterial isolates
5
6
7
18
1
2
Table 2. Organism distribution between <5 mL groups and ≧5 mL groups
organism
<5 mL, number (%)
≧5 mL, number (%)
Staphylococcus aereus
45 (11.8)
32 (9.4)
Streptococcus spp.
29 (7.6)
28 (8.3)
Enterococcus spp.
6 (1.6)
2 (0.6)
Enterobacteriaceae spp.
167 (44.0)
130 (38.3)
Glucose non ferment spp.
38 (10.0)
26 (7.7)
Yeast
5 (1.7)
2 (0.6)
Anaerobe
17 (4.5)
8 (2.3)
Contaminant
59 (15.5)
100 (29.5)
Multiple isolates
14 (3.7)
11 (3.2)
Total
380
339
spp:species
3
19
1
2
Table 3. Detection time of organisms in <5 mL group and ≧5 mL groups
organism
<5 mL
≧5 mL
Staphylococcus aereus
12.1hr
14.9 hr
Streptococcus spp.
12.6 hr
10.3 hr
Enterococcus spp.
12.7 hr
13.1 hr
Enterobacteriaceae spp.
11 hr
7.2 hr
Glucose non ferment spp.
19.4 hr
20.7 hr
Yeast
--
43.3 hr
Anaerobe
62.7 hr
80.3 hr
Contaminant
34.3 hr
28.8 hr
Multiple isolates
12.8 hr
8.6 hr
Average
17.9 hr
17.6 hr
spp:species
3
20