|Year : 2019 | Volume
| Issue : 12 | Page : 1693-1697
Blood culture results at a research and training hospital and the importance of training
FM Sezgin1, UT Babaoglu2
1 Department of Medical Microbiology, Ahi Evran University School of Medicine, Kirsehir, Turkey
2 Department of Public Health, Ahi Evran University School of Medicine, Kirsehir, Turkey
|Date of Submission||13-Nov-2018|
|Date of Acceptance||29-Jul-2019|
|Date of Web Publication||3-Dec-2019|
Dr. F M Sezgin
Department of Medical Microbiology, Ahi Evran University School of Medicine, Kirsehir 40100
Source of Support: None, Conflict of Interest: None
| Abstract|| |
Objectives: This study aims to measure the level of knowledge of the nurses who collect blood cultures at our hospital, and after providing the necessary training, evaluate the distribution of microbial growth and rate of contamination in blood cultures that are referred to our laboratory during a 1-year term. Methods and Materials: A survey was conducted to assess the level of knowledge regarding blood culture acquisition with the participation of 99 nurses at our hospital in October 2017. Blood cultures sent to our laboratory during 2017 May-October were retrospectively evaluated in terms of their results, contamination rates, and number of bottles. Taking survey results into account, monthly trainings were provided to the nurses for 6 months starting from October 2017, and blood culture results and error rates were investigated prospectively. Results: It was determined from the survey results that the level of knowledge regarding the need to wipe the rubber septum of the blood culture bottle with alcohol prior to adding the blood sample (23.2%) and definition of a blood culture set (25.3%) were quite low. It was found that while the contamination rate prior to training was 6.4%, it fell to 3.7% after training, and although the rate of single-bottle cultures was 6.3% before training, it decreased by 2.0%. Conclusions: Standardizing blood culture acquisition with the provided training will produce maximal benefit for every laboratory in terms of cost and workload.
Keywords: Blood culture, contamination, training
|How to cite this article:|
Sezgin F M, Babaoglu U T. Blood culture results at a research and training hospital and the importance of training. Niger J Clin Pract 2019;22:1693-7
| Introduction|| |
In spite the recent advances in diagnostic methods, blood cultures prevail as the most reliable method in the diagnosis of bacteremia and fungemia. Numerous laboratories have been using the automatized and manual blood culture systems at higher rates in the last 20 years. To utilize these methods with most benefit and least cost, clinicians and nurses must be as informed as laboratory personnel.
The evaluation of blood cultures obtained for the diagnosis of bloodstream infections may sometimes pose problems. Clinical evaluation of blood culture results is dependent on many factors associated with all the various stages of the procedure such as the number of collected blood cultures, time of collection, amount of blood collected, technique for drawing blood, skin antisepsis, and whether the person who obtained the sample is well-trained personnel. As preanalytical processes require the participation of units external to the laboratory, they are relatively more challenging to standardize, and it was shown in various publications that most errors originated during this process. Particularly in situ ations where skin antisepsis and disinfection of the bottle tops are not properly done, contamination rates demonstrate increases owing to the rich medium content of the blood culture bottles. Correct interpretation of positivity in blood cultures is a significant problem for the clinical microbiology laboratory, especially when the microorganisms that manifested growth are among those that could be contaminants and when multiple samples could not be obtained for various reasons. Considering that variables such as the genus of the microorganism, microbial load in the sample, variable growth times of microorganisms, and the duration that the sample was kept in room temperature can affect the output time, it is clear that producing reports from these samples is not easy. Fast and accurate interpretation of blood culture results cause the pathogens to be identified in the shortest possible time, the treatment to be guided by accurate reports of antibiotic sensitivity tests, and hence, mortality to decrease.
This study is designed to determine the factors that affect our blood culture contamination rates. Our aim is to measure the level of knowledge of the nurses who obtain blood cultures at our hospital, and after providing the necessary training, evaluate the distribution of microbial growth in the blood cultures that are referred to our laboratory during a 1-year term and associated rates of contamination.
| Materials and Methods|| |
This study was carried out in accordance with the Declaration of Helsinki Guidelines for Good Clinical Practice and was approved by the Local Ethical Committee of the Ahi Evran University Medical Faculty (2017-10/94). The hospitals' authorities also permitted to conduct the study. A survey assessing the level of knowledge regarding blood culture acquisition was conducted with the participation of 99 nurses at the Ahi Evran University Research and Training Hospital in October 2017. Blood cultures sent to our laboratory during 2017 May-October were retrospectively evaluated in terms of their results, contamination rates, and number of bottles. Taking survey results into account, monthly trainings were provided to the nurses for 6 months starting from October 2017, and blood culture results and error rates were investigated prospectively.
In this study, as the method of measurement, a question form to assess the level of knowledge regarding blood culture acquisition was developed in accordance to the literature by the researchers. The question form is composed of 32 items. The question form consists of two sections, which include a 7-item socio-demographic characteristics section, and a 25-item section that is directed at assessing the level of knowledge regarding blood culture collection and practical habits and attitudes. The level of knowledge regarding blood culture collection and practical habits and attitudes section was prepared according to the blood culture collection manual.
For descriptive data, the units used were number (n), percentage (%), and/or mean ± standard deviation values. Categorical variables were evaluated with Chi-square analyses. The data were evaluated using the SPSS 23.0 packaged program. A P value below 0.05 was considered statistically significant.
The catheter culture samples were excluded study. Blood samples were taken from peripheral veins by venous puncture. As explained in the training, prior to venous puncture, the rubber head of the blood culture flask was requested to be disinfected with 70% alcohol. After selecting the appropriate site for a venous puncture, a tourniquet is applied and the vein is palpated, after cleansing and drying the skin with 70% isopropyl alcohol/ethyl alcohol, 10% povidone-iodine is applied from the center to the periphery.
Blood samples were collected in “Bactec Ped Plus” aerobic bottles for pediatric patients and in “Bactec-Plus” aerobic bottles (Becton-Dickinson, USA) for adult patients. Samples were introduced to the BACTEC 9120 automatized system and incubated in the device for 5 days. Bottles that produced positive signals were inoculated to blood agar, eosin-methylene blue, and chocolate agar media; and simple staining with methylene blue and gram staining were performed. Microorganisms were identified using manual methods and the Vitek-2 automatized system (bioMerieux, France).
| Results|| |
According to the results from the survey assessing the level of knowledge of the nurses on blood culture acquisition, the mean age of the participants was 34.78±6.48 and 82.8% were female. Of the nurses who participated, 48.5% had a Bachelor's degree. Nurses reported having 13.85±7.78 years of vocational experience and having worked at the same institution for 7.99 ± 6.16 years. Approximately, 14.1% of the nurses were working in intensive care, 12.1% in general surgery, and 12.1% in coronary intensive care unit (CICU) [Table 1]. The nurses were determined to have worked as clinical nurses for 10.15±6.91 years. About, 56.6% of the nurses reported having received blood culture collection training before. Of these nurses, 51.5% were determined to have received the training after their graduation. However, 93.9% of the participants stated they were aware of the importance of blood cultures for diagnosis, and 89.9% expressed that blood culture was not a routine test. They reported getting most of their information on obtaining blood cultures from their more experienced colleagues or the physicians who ordered the test. Nurses who participated in the study reported contacting the laboratory most frequently for the reason of getting information regarding the status of microbial growth in blood cultures. When we asked the nurses about when they obtained blood cultures, 54.5% responded that they obtained them as soon as the physician ordered the test. When asked about whether they obtained blood cultures from patients under antibiotic therapy, 31.3% reported not obtaining cultures and 62.6% obtaining cultures prior to the start of the therapy. The state of knowledge of the nurses on obtaining blood cultures has been presented in [Table 2]. The nurses reported the amount of sample collected for blood culture as 7.94±2.88 ml on average. They reported that 1.93±0.42 bottles needed to be obtained while collecting blood cultures.
|Table 1: Socio-demographic properties of nurses who participated in the study|
Click here to view
|Table 2: The participants' level of knowledge regarding blood culture acquisition|
Click here to view
Of the 2,969 blood culture bottles that reached our laboratory during a 1-year period, 382 (12.9%) manifested growth, and the strain was considered to be a pathogen in 216 (57%) and to be a contaminant in 166 (43%) of these [Table 3]. It was found that while the contamination rate prior to training was 6.4%, it fell to 3.7% after training. This was statistically significant (P< 0.001). The agents that were most frequently isolated in our study were coagulase-negative staphylococci (CNS), and they were followed by Escherichia More Details coli (E. coli). The distribution of microbial growth has been presented in [Table 4]. Single-bottle samples were obtained from 121 (4%) patients. Although the rate of single-bottle samples was 6.3% before training, it decreased by 2.0% in the end of the training. This was statistically significant (P< 0.001). [Figure 1] demonstrates the graphical change in contamination and single-bottle sample rates by each month.
|Table 3: Numbers and rates of inspected blood cultures, reproduction, and contamination|
Click here to view
|Table 4: Distribution of microorganisms that manifested growth in blood cultures over a 1-year period|
Click here to view
|Figure 1: Graph demonstrating the change in contamination and single-bottle sample number by each month|
Click here to view
| Discussion|| |
Considering the reality of the financial terms that surround our country, health expenses must be carefully monitored, and costs must be given importance. Rapid identification of the pathogenic bacteria in the blood culture and initiation of the appropriate antibiotic therapy is extremely important with regard to mortality. However, when skin antisepsis is not performed properly, and especially when samples are collected in a single bottle, contaminant bacteria reproduce, and pathogen-contaminant discrimination becomes more difficult. As a result, an unnecessary increase in workload and cost is encountered. The negligent behavior manifested by the personnel working at clinics calls for active inclusion of clinical microbiologists in the training of the personnel. We aimed to minimize mistakes through the training we provided after assessing the knowledge levels of nurses with the survey study we conducted.
According to the results from our survey, responses by the personnel who obtain blood cultures at clinics presented high rates regarding the importance blood cultures in diagnosis (93.9%), the importance of the sample amount (96%), the importance of the number of bottles (83.8%), the importance of obtaining samples from different veins (87.9%), obtaining samples after disinfecting the skin and waiting an appropriate time (80%), and the importance of keeping samples in room temperature if the samples need to be stored prior to transfer (86.9%). However, it came to our attention that rates associated with certain variables that could result in an increase in contamination rates, such as knowing that the bottle top must be wiped with alcohol before collecting the sample in the bottle (23.2%), being able to define a blood culture set (25.3%), and knowing the importance of the age and weight of the patient for determining the amount of sample to be collected (12.1%) were quite low, and these issues were given priority in training.
The acceptable contamination rate in automatized blood culture systems has been determined to be ideally below 3%. Studies report contamination rates as high as 14%., Similarly, we have determined contamination rates of 6.4% prior to training and found that these rates regressed to 3.7% after training. Contamination rates are closely related to the technique for obtaining samples, site (catheter or venous puncture), and personnel that draws blood from the patients. Lower contamination rates are found in blood cultures obtained by experienced personnel and specialized blood culture teams. After the survey, training was given every month for the continuation of the improvement in a 6-month period. However, there were fluctuations in the graph showing monthly contamination number owing to external factors such as annual leave, personnel relocation, and internship students [Figure 1]. In a study by Herwaldt and colleagues, a contamination rate as low as 1.2% could be obtained when samples were collected with defined blood culture procedures and by trained phlebotomist.
As a limitation of our study, we stated that the survey was conducted only once. The study could not be repeated at the end.
It is important to obtain absolutely two blood culture bottles from different veins both because rates of microorganism identification are low with single blood culture and because in cases that CNS growth is detected a single blood culture is not adequate for evaluating whether it is pathogenic or contamination. While the rate of single-bottle samples sent to our laboratory was 6.3% before training, we found that this rate decreased by 2.0% after training. Therefore, the associated cost was reduced by preventing both unnecessary laboratory workload and unnecessary microorganism identification and antibiogram work.
A review of the literature reveals that the most commonly isolated microorganisms in patients with positive blood cultures are gram-positive cocci (CNS, S. aureus, and Enterococci), enteric gram-negative bacilli (E. coli, Klebsiella, and Enterobacter spp.), and yeasts., In recent years, gram-positive cocci, in particular, have been detected more frequently as pathogens. Among agents isolated from blood cultures, rates of gram-positive bacteria have been reported in different studies as 64.1%, 80%, and 68.8%. Various studies done in our country reported CNS as the most commonly isolated microorganism.,,,
Safak and colleagues investigated the results of blood cultures obtained from inpatients between the years 2010 and 2015 and determined the most frequently isolated microorganisms as CNS (35.6%), S. aureus (27.8%), and E. coli (10.8%). Similarly, in our study, when we only considered laboratory findings, CNS was the most frequently isolated bacteria with a rate of 50%, followed by E. coli (18%) and enterococci (12.5%). E. coli was the most frequently isolated agent among gram-negative bacteremia agents, paralleling numerous other studies. In accordance with the literature, long-term hospitalization and the resulting increase in invasive procedures were found to cause infections induced by these pathogens.
| Conclusion|| |
All hospitals are required to closely monitor blood culture results and contamination rates, particularly owing to quality standards. The evaluation of blood cultures obtained for the diagnosis of bloodstream infections continues to pose problems. The clinical evaluation of culture results can be affected by multiple factors at almost all stages of the procedure, which include the time of blood collection, technique for drawing blood, volume, skin antisepsis, whether the person who obtained the sample is well-trained personnel, and the number of cultures. Standardization of these factors with continual training will provide maximal benefit to every laboratory in terms of cost and workload.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Hall KK, Lyman JA. Updated review of blood culture contamination. Clin Microbiol Rev 2006;9:788-802.
Çiçek A, Kuzucu Ç, Durmaz B. Kan kültür sonuçlarının deǧerlendirilmesinde etkili olan faktörler. İnönü Üniv Tıp Fak Derg 2005;12:277-80.
Lippi G, Guidi GC. Risk management in the preanalytical phase of laboratory testing. ClinChem Lab Med 2007;45:720-7.
Ntusi N, Aubin L, Oliver S, Whitelaw A, Mendelson M. Guideline for the optimal use of blood cultures. S Afr Med J 2010;100:839-43.
Başustaoǧlu A, editor. Kan kültürü uygulama klavuzu. Ankara: Türk Mikrobiyoloji Cemiyeti 2013.
Kuzucu Ç, Ayan M, Durmaz B. Üç aylık periyoda kan kültür sonuçlarının deǧerlendirilmesi. İnönü Üniv Tıp Fak Derg 2001;8:25-8.
Dunne WM, Nolte FS, Wilson ML. In Hindler JA, editor. Blood Cultures III. Curnitech Series, American Society for Microbiology. Curnitech IB: Washington DC; 1997.
Richter SS, Beekmann SE, Croco JL, Diekema DJ, Koontz FP, Pfaller MA,et al
. Minimizing the workup of blood culture contaminants: Implementation and evaluation of a laboratorbased algorithm. J Clin Microbiol 2002;40:2437-44.
Herwaldt LA, Hollis RJ, Boyken LD, Pfaller MA. Molecular epidemiology of coagulase negative staphylococci isolated from immunocompromised patients. Infect Control Hosp Epidemiol 1992;13:86-92.
Çiçek A, Kuzucu Ç, Durmaz R. Bir yıllık sürede kan kültürlerinin klinik epidemiyolojik ve bakteriyolojik yönden prospektif analizi. Flora 2006;11:3744.
Gültekin E, Uyanık MH, Hancı H, Erdil Z, Gelen FN, Çelebi S. Kan kültürlerinden izole edilen nonfermentatif Gram negatif bakterilerin çeşitli antibiyotiklere duyarlılıkları. Ankem Derg 2014;28:79-85.
Ece G. Kan kültüründe üreyen izolatların daǧılım ve antibiyotik duyarlılık profilinin incelenmesi. Haseki Tıp Bülteni 2013;51:151-6.
Gülmez D, Gür D. Hacettepe Üniversitesi İhsan Doǧramacı Çocuk Hastanesi'nde 2000-2011 yılları arasında kan kültürlerinden izole edilen mikroorganizmalar: 12 yıllık deǧerlendirme. J Pediatr Infect 2012;6:79-83.
Çopur Çiçek A, Şentürk Köksal Z, Ertürk A, Köksal E. Rize 82.Yıl Devlet Hastanesi'nde bir yıllık sürede kan kültürlerinden izole edilen mikroorganizmalar ve antibiyotiklere duyarlılıkları. Türk Hij Den Biyol Derg 2011;68:175-84.
Şafak B, Kılınç O. 2010-2015 yılları arasında kan kültürlerinde üreyen mikroorganizmalar ve antibiyotik duyarlılıkları. Klimik Derg 2016;29:61-5.
Balıkçı A, Belas Z, Eren Topkaya A. Kan kültürü pozitifliǧi: Etken ya da kontaminasyon mu? Mikrobiyolji Bul 2013;47:135-40.
[Table 1], [Table 2], [Table 3], [Table 4]