Article, Emergency Medicine

The impact of overcrowding on the bacterial contamination of blood cultures in the ED

Unlabelled imageAmerican Journal of Emergency Medicine (2012) 30, 839-845

Original Contribution

The impact of overcrowding on the Bacterial contamination of blood cultures in the ED

Ching-Chi Lee MD a, Nan-Yao Lee MD a, Ming-Che Chuang MD b, Po-Lin Chen MD a,

Chia-Ming Chang MD a, Wen-Chien Ko MD a,c,?

aDepartment of Internal Medicine, National Cheng Kung University Hospital, Tainan, Taiwan bDepartment of Emergency Medicine, National Cheng Kung University Hospital, Tainan, Taiwan cDepartment of Medicine, National Cheng Kung University Medical College, Tainan, Taiwan

Received 1 April 2011; revised 13 May 2011; accepted 19 May 2011

Abstract

Objectives: This study aims to determine the risk factors associated with the bacterial contamination of blood cultures among adults visiting the emergency department (ED).

Methods: Clinical variables and medical records of adults with bacterial growth of blood cultures in the ED as well as the degree of ED crowding, between August 2007 and July 2008, were prospectively collected.

Results: Of the 11 491 adults who underwent blood culture sampling, the medical records of 558 (4.86%) eligible patients with bacterial growth in their blood cultures were analyzed. Most patients (366, or 3.19%) had true bacteremia, whereas 192 (1.67%) were regarded as contaminated. In multivariate analyses, ED overcrowding (scoring was based on a National Emergency Department Overcrowding Study [NEDOCS] score >=100 points) was independently associated with blood culture contamination (odds ratio [OR], 1.58; P = .04). In contrast, other Medical comorbidities, such as liver cirrhosis (OR, 0.31; P = .02), thrombocytopenia (b100 000/mm3; OR, 0.28; P = .002), or high serum levels of C-reactive protein (N100 mg/L; OR, 0.24; P b .001), were negatively associated with blood culture contamination. On further analysis of the 5 crowding categories as stratified by NEDOCS scores, which included not busy and busy (0-60 points), extremely busy but not overcrowded (60-100), overcrowded (100-140), severely overcrowded (140-180), and dangerously overcrowded (180-200), there was a strong correlation between blood culture contamination rates and the degrees of ED crowding (? = 0.99, P b .001).

Conclusions: Emergency department overcrowding may have an adverse impact on the quality of clinical care, including increasing the risk of blood culture contamination.

(C) 2012

Introduction

* Corresponding author. Department of Internal Medicine, National Cheng Kung University Hospital, 70403, Tainan, Taiwan. Tel.: +886 6 2353535×3596; fax: +886 6 2752038.

E-mail address: [email protected] (W.-C. Ko).

Bacteremia is a life-threatening condition associated with high morbidity and mortality. Indeed, the case-fatality rate is as high as 30% [1,2]. Blood culture sampling is a mandatory diagnostic test used to detect bacteremia and to identify the causative microorganisms for further antimicrobial suscep- tibility testing. However, contaminated blood cultures

0735-6757/$ – see front matter (C) 2012 doi:10.1016/j.ajem.2011.05.026

comprise up to a half of all positive blood cultures in adults [3,4]. Thus, the appropriate interpretation of culture results often requires sophisticated microbiological knowledge.

Contaminated cultures often adversely impact health care and medical expenditure because patients with contaminated blood cultures may undergo unnecessary hospitalizations. Such false-positive results may lead to the unnecessary administra- tion of antimicrobial agents and additional microbiological studies or supplementary investigations, such as serum C-reaction protein or procalcitonin level [3,5], to assist in ascertaining the clinical significance of the blood culture results. Numerous studies have reported that emergency depart- ment (ED) overcrowding is a common issue in many hospitals in both the United States and Taiwan and that it threatens patient safety and public health [6,7]. Several investigators have described the relationship between ED overcrowding and clinical care. Indeed, the data suggest that ED overcrowding is associated with delays in clinical diagnosis and appropriate management as well as poor patient outcomes [8]. However, in the literature, no studies have explored the association between ED overcrowding and blood culture contamination. Thus, the aims of this study were to analyze the clinical characteristics and outcomes of patients with Bloodstream infections and those with contaminated blood cultures and to investigate the factors

related to blood culture contamination in the ED.

Methods

Study design

This prospective study was conducted at a teaching hospital with approximately 1000 beds in southern Taiwan between August 2007 and July 2008. This study was approved by the hospital institutional review board. Adults who were at least 18 years old with bacterial growth in blood cultures collected in the ED were included in the study, and their medical records were reviewed. The degree of ED overcrowding was measured using the National Emergency Department Overcrowding Study (NEDOCS) score [9].

Clinical characteristics, Pittsburgh bacteremic severity score, comorbidities, and laboratory parameters in the ED were collected. The NEDOCS score was calculated routinely every 2 hours, and the NEDOCS score corresponding to a positive blood culture was the value closest to the timing of the blood sampling. Emergency department visits made by the same patient were considered distinct if the visits were separated by at least 2 weeks or if the ED visits were related to different causes.

For this hospital, common procedures of skin disinfection before blood sampling were completed in the following manner. The venipuncture sites were initially disinfected with 75% alcohol; cleansing of the skin was performed in concentric circles, beginning at the center of the site. The site

was then disinfected with 10% povidone-iodine, again in concentric circles. The disinfected areas were allowed to air dry for at least 1 minute. The above procedures were repeated 3 times before penetrating the skin. Before injection into the culture bottles, the tops of the bottles were cleaned with 75% alcohol.

After skin disinfection, blood sampling was performed by the nurses, and 2 sets of blood cultures were collected from different peripheral veins or arteries 30 minutes apart. One set of blood cultures was routinely composed of a bottle for aerobic culture and another for anaerobic culture. Approxi- mately 5 to 8 mL of blood was collected in each bottle. The culture bottles were immediately transported to the hospital laboratory, where they were loaded into the BACTEC 9240 system (Becton Dickinson and Co, Franklin Lakes, NJ) and incubated for 5 days or until the instrument indicated bacterial growth. The culture bottles with bacterial growth underwent Gram staining, and the attending clinicians were informed of the Gram staining results by microbiologic technicians. At the same time, the contents of the bottles were subcultured onto bioplates with blood agar (Trypticase soy agar II 5% sheep blood; Becton Dickinson and Co), Levine EMB agar (Becton Dickinson and Co), chocolate agar, and CDC anaerobic blood agar (Becton Dickinson and Co) for further identification. Biochemical tests and automatic identification systems were used for final identification.

Definitions

Polymicrobial bacteremia was defined as the isolation of more than 1 microbial species from 2 sets of blood cultures. The severity of bloodstream infections in the ED was measured using the Pittsburgh bacteremia score, which is a validated scoring system based on vital signs, mental status, mechanical ventilation, and the presence of cardiac arrest [10]. The degree of ED overcrowding was measured using the NEDOCS score [9]. Blood cultures with potential contaminant pathogens, such as coagulase-negative Staphylococcus, Micrococcus, Propio- nibacterium acnes, Peptostreptococcus, or Bacillus species, were regarded as contaminated based on the previously described criteria [11]. Polymicrobial cultures with a mixture of contamination isolates and true pathogens were regarded as contaminated samplings. The definitions of comorbidities (such as malignancy, diabetes mellitus, hypertension, old stroke, Chronic renal insufficiency, liver cirrhosis, and coronary artery disease) were as previously described [12]. Types of infection were determined to be pneumonia, urinary tract infection, Wound infection, soft-tissue infection, intraab- dominal infection, or primary bloodstream infection, which were based on definitions by Centers for Disease Control and Prevention [13].

Statistical analysis

Statistical analyses were performed using the Statistical Package for the Social Sciences for Windows, Version

old age, >=65 y

209

(57.10)

119 (62.96)

.18

Sex, male

185

(50.55)

105 (55.56)

.26

Recent hospitalization within 1 mo

67

(18.30)

34 (17.99)

.92

Transfer from

Other hospitals

26

(7.10)

14 (7.40)

.89

Chronic health care facilities

21

(5.73)

12 (6.34)

.77

Major comorbidities

Hypertension

189

(51.64)

92 (48.68)

.50

Diabetes mellitus

133

(36.34)

48 (25.40)

.009

Malignancy

86

(23.50)

29 (15.34)

.02

Old stroke

52

(14.21)

42 (22.22)

.01

Chronic renal insufficiency

58

(15.85)

22 (11.64)

.18

Liver cirrhosis

45

(12.30)

6 (3.17)

b.001

Coronary artery diseases

33

(9.02)

17 (9.08)

.99

Pittsburgh bacteremia score >=4 points in the ED

67

(18.31)

47 (24.87)

.07

Laboratory data in the ED a

Leukocyte N9000/mm3

236/360

(65.56)

106/174 (60.92)

.29

Platelet b100 000/mm3

65/354

(18.36)

8/170 (4.71)

b.001

Blood urea nitrogen N20 mg/dL

194/357

(54.34)

74/167 (44.31)

.03

Serum creatinine N1.5 mg/dL

109/356

(30.62)

37/167 (22.16)

.04

C-reactive protein N100 mg/L

117/327

(35.78)

20/156 (12.82)

b.001

Serum glucose N200 mg/dL

97/335

(28.96)

36/162 (22.22)

.11

serum albumin b3.5 g/dL

60/93

(64.52)

17/35 (48.57)

.10

Overcrowded ED b at blood culture sampling

267

(72.95)

155 (82.01)

.01

a Not all patients had the indicated laboratory data.

b National Emergency Department Overcrowding Study score >=100 points.

14.0 (SPSS, Chicago, Ill). Continuous variables were expressed as the mean +- SD and compared by the Student t test. Categorical variables, expressed as the numbers and percentages, were compared by the ?2 or Fisher exact test. The variables with a P value less than .1 on univariate analysis contributed to the stepwise, backward logistic regression model. The correlation between 2 continuous variables was analyzed by the Pearson correla- tion. A P value less than .05 was considered to be statistically significant.

Table 1 Comparisons of demographic data, clinical characteristics, and laboratory parameters in adults with true bacteremia and those with contaminated sampling drawn at the ED after exclusion of 3 patients with mixed isolates of true bacteremic and contaminated isolates

Characters

Episode no. (%)

True bacteremia, n = 366

P

Contamination, n = 189

Results

Factors related to the contamination of blood cultures collected in the ED

The annual number of ED visitors was approximately 66 000. The mean value (+-SD) of the hospital-based NEDOCS score was 128.6 (+-41.3), and the mean duration of stay in the ED before admission was 16.8 (+-6.9) hours during the study period. During the 1-year study period, of the 11 491 adults with blood culture samplings, 558

(4.86%) patients who had bacterial growth in their blood cultures were eligible in the present study. Most (366 patients, or 65.60%) had true bacteremia, and the blood cultures of 189 (33.87%) patients were contaminated, whereas 3 patients (0.53%) had mixed true bacteremic and contaminated isolates. Therefore, the contamination rate of blood cultures during the study period was 1.67% (192/11 491).

Of the total 208 isolates from the 192 adults with contaminated blood cultures, the most common microorgan- ism was coagulase-negative staphylococci (142 isolates, or 68.20%), followed by Bacillus species (34, or 16.35%),

Propionibacterium acnes (13, or 6.25%), Micrococcus

species (10, or 4.80%), Peptostreptococcus species (4, or

1.92%), Clostridium perfringens (3, or 1.44%), and ?

hemolytic streptococci (2, or 0.96%).

Of the 558 patients, most (485, or 86.92%) visited the ED from the community, 33 (5.91%) were transferred from long- term health care facilities, and 40 (8.78%) were transferred from other hospitals. Their mean age was 64.7 years, and 291 (52.15%) were men. Major comorbidities included hyper- tension (283 patients, or 50.72%), diabetes mellitus (182, or

32.62%), malignancy (116, or 20.79%), stroke (96, or

17.20%), chronic renal insufficiency (81, or 14.52%), liver

Old age, >=65 y

50

(51.02)

116

(62.37)

.06

Sex, male

62

(63.26)

103

(55.38)

.20

Recent hospitalization within 1 mo

18

(18.37)

34

(18.28)

.98

Transfer from

Other hospitals

12

(12.24)

14

(7.53)

.19

Chronic health care facilities

6

(6.12)

11

(5.38)

.94

Polymicrobial isolates

4

(4.08)

13

(6.99)

.32

Major comorbidities

Hypertension

47

(47.96)

90

(48.39)

.94

Diabetes mellitus

35

(35.71)

47

(25.29)

.06

Chronic renal insufficiency

25

(25.51)

21

(11.29)

.002

Malignancy

18

(18.37)

30

(16.13)

.63

Old stroke

18

(18.37)

41

(22.04)

.46

Liver cirrhosis

18

(18.37)

6

(3.22)

b.001

Coronary artery diseases

14

(14.29)

17

(9.14)

.18

Pittsburgh bacteremia score >=4 points in the ED

17

(17.35)

48

(25.81)

.10

Laboratory data in the ED a

Leukocyte N9000/mm3

72/98

(73.47)

103/171

(60.23)

.02

Platelet b100 000/mm3

12/97

(12.37)

9/167

(5.39)

.04

Blood urea nitrogen N20 mg/dL

60/97

(61.86)

73/164

(44.51)

.007

Serum creatinine N1.5 mg/dL

36/98

(36.73)

38/164

(23.17)

b.001

C-reactive protein N100 mg/L

38/88

(43.18)

21/153

(13.73)

b.001

Serum glucose N200 mg/dL

28/94

(29.78)

36/159

(22.64)

.20

Serum albumin b3.5g/dL

16/25

(64.00)

18/35

(51.43)

.33

a Not all patients had the indicated laboratory data.

cirrhosis (51, or 9.14%), and coronary artery disease (50, or 8.96%).

Table 2 Demographic data, clinical characteristics, and laboratory parameters in the ED of 284 adults with Gram-positive cocci growing in the blood cultures, either true bacteremia or contamination

Characters

Patient no. (%)

True bacteremia, n = 98

P

Contamination, n = 186

Differences between bacteremic patients and those with contaminated blood cultures in terms of demographic data, clinical characteristics, laboratory parameters, and Pittsburgh bacteremia score were examined by univariate analysis, as shown in Table 1. Of the clinical characteristics, stroke (22.2% vs 14.2%; P = .01) and ED overcrowding at the time of blood culture sampling (NEDOCS >=100 points; 82.0% vs 73.0%; P = .01) were more often noted in patients with contaminated blood cultures compared with those with true bacteremia.

In the multivariate analysis, ED overcrowding (NEDOCS

>=100 points) was independently associated with contami- nated blood cultures (OR, 1.58; 95% CI, 1.01-2.59; P = .04). In contrast, the presence of thrombocytopenia (b100 000/ mm3; OR, 0.28; 95% CI, 0.12-0.63; P = .002), high serum

C-reactive protein (N100 mg/L; OR, 0.24; 95% CI, 0.11- 0.50; P b .001), and comorbidities with liver cirrhosis (OR, 0.31; 95% CI, 0.11-0.81; P = .02) were negatively associated with contaminated blood cultures.

Contamination in Gram-positive cocci bacteremia

Because most of the contaminated blood cultures were caused by Gram-positive cocci (158/208 isolates, or 76.0%),

it is useful to identify risk factors for true bacteremic episodes among patients with the growth of Gram-positive cocci in blood cultures. Such information may be useful in minimizing the unnecessary use of antimicrobial agents in patients with blood culture contamination. After the exclusion of 274 patients with the growth of Gram-negative organisms in their blood cultures, there were 284 adults with Gram-positive cocci in their blood cultures. Demographic data, clinical characteristics, and laboratory parameters were available for 98 (34.5%) patients with true Gram-positive cocci bacteremia and 186 (65.5%) with contaminated blood cultures; these data were compared and displayed in Table 2. Based on the univariate analysis, compared with patients with true bacteremia, the patients with contaminated blood cultures less frequently had chronic renal insufficiency (11.3% vs 25.5%, P = .002), liver cirrhosis (3.2% vs

18.4%, P b .001), leukocytosis (N9000/mm3, 60.2% vs

73.5%, P = .02), thrombocytopenia (b100 000/mm3, 5.4% vs 12.4%, P = .04), high blood urea nitrogen (N20 mg/dL, 44.5% vs 61.9%, P = .007), high serum creatinine (N1.5 mg/dL, 23.2% vs 36.7%, P b .001), and high C-reactive protein (N100 mg/L, 13.7% vs 43.2%, P b .001). However, in the multivariate regression analysis, of patients with Gram-positive cocci bacteremia in the ED, old age (>=65 years) was significantly associated with contamination (OR, 1.79; 95% CI, 1.08-3.20; P = .04). Nevertheless, the

presence of liver cirrhosis (OR, 0.09; 95% CI, 0.03-0.26; P b

.001) or renal disease (OR, 0.32; 95% CI, 0.12-0.84; P = .02)

in the ED was negatively associated with contamination among patients with Gram-positive cocci bacteremia.

Blood culture contamination and ED crowding

Adults with positive blood cultures in the ED were divided into 6 categories using the following stratification of NEDOCS scores: not busy (0-20 points), busy (20-60 points), extremely busy but not overcrowded (60-100 points), overcrowded (100-140 points), severely over- crowded (140-180 points), and dangerously overcrowded (180-200 points). Because of the limited case numbers in the not-busy and busy groups, the 2 categories were merged into

1. Demographic data, major comorbidities, bacteremic features, and laboratory parameters for the 5 ED crowding groups were compared (Table 3). In the univariate analysis, only the blood culture contamination rates (P = .02), high Pittsburgh bacteremic scores (>=4 points) (P = .04), underlying illness with liver cirrhosis (P = .02), and elevated C-reactive protein (N100 mg/L) (P = .01) differed among the 5 ED crowding groups. However, only the blood culture contamination rate increased consistently with the degree of

Table 3 Demographic data, comorbidities, and laboratory parameters of 558 adults with positive blood cultures in the ED, stratified into different degrees of ED crowding based on the NEDOCS score

ED crowding (? = 0.99, P b .001), as shown in Fig. 1. The correlation between ED crowding and elevated C-reactive protein (? = -0.53, P = .34), high Pittsburgh bacteremia score (? = 0.28, P = .64), or the presence of liver cirrhosis (? = -0.71, P = .10) was poor.

Discussion

Blood culture contamination is a common clinical problem and often leads to both inappropriate treatment and an adverse impact on Health care costs. Therefore, several strategies have been recommended to decrease the rates of contaminated blood cultures, including the follow- ing: the use of disinfectants, educational interventions, sampling from separate venipuncture sites, appropriate preparation of culture bottles, the use of the outmoded double-needle technique, reliance on a well-trained phlebot- omy team, and the use of commercial culture kits [5,14]. However, if focusing on the ED, it is challenging to prevent contamination because of inadequate inpatient capacity, a high severity of illness, the rapid turnover of staff, the lack of ongoing training, and the high workload [15]. There have been only a few studies that have proposed measures to

Variables

Patient no. (%)

P

Not busy and

Extremely busy but not

Overcrowded,

Severely

Dangerously

busy, n = 25

overcrowded, n = 106

n = 214

overcrowded,

overcrowded,

n = 141

n = 72

NEDOCS score

0-60

60-100

100-140

140-180

180-200

Old age, >=65 y

15 (60)

55 (51.88)

123 (57.48)

94 (66.67)

41 (56.94)

.20

Sex, male

14 (56)

45 (42.45)

113 (52.80)

78 (55.32)

41 (56.94)

.28

bacteremic isolates

Polymicrobial

0 (0)

11 (10.38)

17 (7.94)

12 (8.51)

8 (11.11)

.48

Contamination

5 (20)

29 (27.36)

68 (31.78)

57 (40.42)

33 (45.83)

.02

Major comorbidities

Hypertension

13 (52)

50 (47.17)

114 (53.27)

69 (48.94)

37 (51.39)

.86

Diabetes mellitus

7 (28)

40 (37.74)

69 (32.22)

49 (34.75)

17 (23.61)

.34

Malignancy

9 (36)

24 (22.64)

38 (17.76)

28 (19.86)

17 (23.61)

.25

Old stroke

2 (8)

11 (10.38)

35 (16.36)

32 (22.70)

13 (18.06)

.12

Chronic renal insufficiency

3 (12)

21 (19.81)

35 (16.36)

13 (9.22)

10 (13.89)

.13

Liver cirrhosis

3 (12)

16 (15.09)

22 (10.28)

5 (3.55)

5 (6.94)

.02

Coronary artery diseases

3 (12)

10 (9.43)

18 (8.41)

10 (7.09)

9 (12.50)

.71

Laboratory examination in the ED Leukocyte N9000/mm3

14 (56)

77 (72.64)

127 (59.35)

82 (58.16)

43 (59.72)

.10

Platelet b100 000/mm3

6 (24)

16 (15.09)

25 (11.68)

18 (12.77)

9 (12.50)

.45

Blood urea nitrogen N20 mg/dL

9 (36)

51 (48.11)

105 (49.07)

74 (52.48)

31 (43.06)

.50

Serum creatinine N1.5 mg/dL

4 (16)

33 (31.13)

60 (28.03)

30 (21.28)

21 (29.17)

.23

Serum glucose N200 mg/dL

3 (12)

28 (26.41)

54 (25.23)

34 (24.11)

14 (19.44)

.77

Serum albumin b3.5 g/dL

3 (12)

13 (12.26)

32 (14.95)

18 (12.77)

12 (16.67)

.98

C-reactive protein N100 mg/L

16 (64)

73 (68.87)

135 (63.08)

78 (55.32)

45 (62.50)

.01

Pittsburgh bacteremia score >=4 points

3 (12)

12 (11.32)

146 (68.22)

36 (25.53)

19 (26.39)

.04

in the ED

Fig. 1 Contamination rates among adults with positive blood cultures in the ED with respect to the varying degrees of ED crowding, as stratified by the NEDOCS score. A strong correlation between the contamination rate and the degree of ED crowding was demonstrated (? = 0.99, P b .001).

decrease the rates of contaminated blood cultures in the ED; these measures include the use of simple blood culture sampling kits and the use of phlebotomists [14,15].

Emergency department overcrowding is a well-known systemic problem that creates a high-risk environment for Medical errors [7]. One of the main findings in the present study is the strong link between the extent of ED overcrowding and the contamination rate of blood cultures, which is, to date, the first study to document this. Of note, most clinical variables available at the ED, including major comorbidities and laboratory parameters, were similar between these ED overcrowding categories, with the exception of serum levels of C-reactive protein and the Pittsburgh bacteremia score. However, there was no trend in the proportion of high C-reactive protein levels and high Pittsburgh bacteremic scores between the 5 categories.

In general, the blood culture contamination rate varied widely from as low as 0.6% to greater than 6%, depending on the technique used, the patient population, and the technicians [5]. Although few studies have emphasized the blood culture contamination rate in the ED, the reported rates range from 0.8% to 5.4% [3,16,17]. The contamination rate in our study period was 1.67% and is compatible with previous reports. Moreover, similar to previous reports [3,5], coagulase- negative staphylococci was the most common contaminant.

In addition to ED overcrowding, among ED patients with a bacterial growth on blood culture, liver cirrhosis, the presence of thrombocytopenia, and elevated levels of serum C-reactive protein were factors that were negatively associated with contamination of blood cultures. Such an

association is unique and has not been reported in the English literature. To our knowledge, Cirrhotic patients have often been associated with bacteremic episodes [18,19]. Moreover, several laboratory abnormalities, such as high serum levels of C-reactive protein and thrombocytopenia, are significantly associated with bacteremic episodes, especially bacteremia caused by Gram-negative pathogens [20-22]. In our study population, Gram-negative bacilli were the major pathogens; thus, these laboratory findings independently associated with Gram-negative bacillary bloodstream infections may have contributed to our findings. If ED clinicians were informed of bacterial growth in blood cultures of adult visitors, such laboratory data may be helpful in discriminating true bloodstream infection from contamination, before the microbiological results became available. If the condition is confined to those with Gram-positive cocci on blood cultures, some patient characteristics, such as old age and underlying liver or renal disease, may be useful in making decisions concerning further management.

Despite our finding of an association between blood culture contamination and ED overcrowding, such a result should be interpreted cautiously in light of several limitations inherent in the study design. First, to evaluate the degree of ED overcrowding, the NEDOCS score, a previously validated scale, was prospectively collected. Although the NEDOCS value was not measured at the sampling time of the blood cultures, the scores were measured every 2 hours and the one closest in proximity to the sampling time was adopted. The discrepancy between the actual and adopted NEDOCS scores was presumed to be negligible. Second,

although the patients were enrolled for a 1-year period, the study was conducted at the ED of a tertiary hospital, and the conclusions may not be generalizable to all EDs. Third, contamination rates among blood cultures with bacterial growth in the ED were low in our population. In this study, it was difficult to define the degree of ED overcrowding for each blood culture, and therefore, the true contamination rate of blood cultures obtained in the ED and its correlation with ED crowding cannot be determined. Instead, the contamination rates in adults with blood cultures with bacterial growth were calculated to indirectly reflect the likelihood of contamination in blood cultures obtained in the ED in our study.

In the current study, 5 independent factors related to contamination of blood cultures were found. The potentially amendable factor associated with contamination was ED overcrowding. To improve the quality of clinical care and decrease unnecessary health care costs, it is imperative that the impact of ED crowding be diminished.

References

  1. Leibovici L, Greenshtain S, Cohen O, Mor F, Wysenbeek AJ. Bacteremia in febrile patients. A clinical model for diagnosis. Arch Intern Med 1991;151:1801-6.
  2. Bates DW, Pruess KE, Lee TH. How bad are bacteremia and sepsis? Outcomes in a cohort with suspected bacteremia. Arch Intern Med 1995;155:593-8.
  3. Archibald LK, Pallangyo K, Kazembe P, Reller LB. Blood culture contamination in Tanzania, Malawi, and the United States: a microbiological tale of three cities. J Clin Microbiol 2006;44:4425-9.
  4. Thuler LC, Jenicek M, Turgeon JP, Rivard M, Lebel P, Lebel MH. Impact of a false positive blood culture result on the management of febrile children. Pediatr Infect Dis J 1997;16:846-51.
  5. Hall KK, Lyman JA. Updated review of blood culture contamination. Clin Microbiol Rev 2006;19:788-802.
  6. Shih FY, Ma MH, Chen SC, Wang HP, Fang CC, Shyu RS, et al. ED overcrowding in Taiwan: facts and strategies. Am J Emerg Med 1999;17:198-202.
  7. Trzeciak S, Rivers EP. Emergency department overcrowding in the United States: an emerging threat to patient safety and public health. Emerg Med J 2003;20:402-5.
  8. Cowan RM, Trzeciak S. Clinical review: emergency department overcrowding and the potential impact on the critically ill. Crit Care 2005;9:291-5.
  9. Weiss SJ, Derlet R, Arndahl J, Ernst AA, Richards J, Fernandez- Frackelton M, et al. Estimating the degree of emergency department overcrowding in academic medical centers: results of the National ED Overcrowding Study (NEDOCS). Acad Emerg Med 2004;11: 38-50.
  10. Paterson DL, Ko WC, Von Gottberg A, Mohapatra S, Casellas JM, Goossens H, et al. International prospective study of Klebsiella pneumoniae bacteremia: implications of extended-spectrum beta- lactamase production in Nosocomial infections. Ann Intern Med 2004;140:26-32.
  11. Lee CC, Lin WJ, Shih HI, Wu CJ, Chen PL, Lee HC, et al. Clinical significance of potential contaminants in blood cultures among patients in a medical center. J Microbiol Immunol Infect 2007;40: 438-44.
  12. Schellevis FG, van der Velden J, van de Lisdonk E, van Eijk JT, van Weel C. Comorbidity of chronic diseases in General practice. J Clin Epidemiol 1993;46:469-73.
  13. Garner JS, Jarvis WR, Emori TG, Horan TC, Hughes JM. CDC definitions for nosocomial infections, 1988. Am J Infect Control 1988;16:128-40.
  14. Gander RM, Byrd L, DeCrescenzo M, Hirany S, Bowen M, Baughman

J. Impact of blood cultures drawn by phlebotomy on contamination rates and health care costs in a hospital emergency department. J Clin Microbiol 2009;47:1021-4.

  1. Madeo M, Jackson T, Williams C. Simple measures to reduce the rate of contamination of blood cultures in accident and emergency. Emerg Med J 2005;22:810-1.
  2. Sturmann KM, Bopp J, Molinari D, Akhtar S, Murphy J. Blood cultures in adult patients released from an urban emergency department: a 15-month experience. Acad Emerg Med 1996;3:768-75.
  3. Chiu YH, Chen TJ, Chen CT, Lu CC. Positive blood cultures in pediatric emergency department patients: epidemiological and clinical characteristics. Acta Paediatr Taiwan 2005;46:11-6.
  4. Peduzzi P, Shatney C, Sheagren J, Sprung C. Predictors of bacteremia and Gram-negative bacteremia in patients with sepsis. The Veterans Affairs Systemic Sepsis Cooperative Study Group. Arch Intern Med 1992;152:529-35.
  5. Bates DW, Sands K, Miller E, Lanken PN, Hibberd PL, Graman PS, et al. Predicting bacteremia in patients with sepsis syndrome. Academic Medical Center Consortium Sepsis Project Working Group. J Infect Dis 1997;176:1538-51.
  6. Cohen P, Gardner FH. Thrombocytopenia as a laboratory sign and complication of Gram-negative bacteremic infection. Arch Intern Med 1966;117:113-24.
  7. McCabe RE, Remington JS. C-reactive protein in patients with bacteremia. J Clin Microbiol 1984;20:317-9.
  8. Engel A, Mack E, Kern P, Kern WV. An analysis of interleukin-8, interleukin-6 and C-reactive Protein Serum concentrations to predict fever, Gram-negative bacteremia and complicated infection in neutropenic cancer patients. Infection 1998;26:213-21.

Leave a Reply

Your email address will not be published. Required fields are marked *