Systematic Review on Epidemiology of Escherichia coli in Bloodstream Infection of Patients Undergoing Hematopoietic Stem Cell Transplantation

Abstract

Introduction: We aimed to conduct a systematic review of the epidemiology of Escherichia coli in bloodstream infections (BSI) of hematopoietic stem cell transplantation patients. Methods: For a comprehensive search of studies that reported the prevalence of E. coli and antibiotic resistance in bloodstream infections from 2000 to January 1, 2024, databases such as PubMed, EMBASE, Google Scholar, Scopus, and Web of Science were searched. The main keywords used were: Escherichia coli, epidemiology, bloodstream infection, microbial resistance, antibiotic resistance, hematopoietic malignancy, hematopoietic stem cell transplantation. After applying eligibility criteria, and quality assessment of studies, data analysis was done by comprehensive meta-analysis (CMA) software. Results: The prevalence of bacterial bloodstream infections amongst different studies varied between 8-51%. Also, bloodstream infections caused by E. coli varied between 2.5-57%. Prevalence of extended-spectrumβ-lactamases (ESBLs) of Escherichia coli in bloodstream infections varied between 15-80%. As well, the mortality rate caused by Escherichia coli strains in bloodstream infection varied between 6.7-27.3%. Resistance to ciprofloxacin, cefepime, third- and fourth-generation cephalosporins, was reported to be the highest (prevalence of 100%), and the lowest was against amikacin, with a prevalence between 13-38%. Conclusions: The high prevalence of Escherichia coli-related BSI, and subsequent mortality, especially by multidrug resistance and ESBL strains, in patients undergoing hematopoietic stem cell transplantation, requires essential measures to prevent the spread of microbial resistance.

Introduction

Hematological malignancies such as lymphatic and myeloid tumors are caused by disruption of normal hematopoietic function which include leukemia, non-Hodgkin lymphoma, multiple myeloma, and Hodgkin lymphoma [1]. Overall, the occurrence of hematologic malignancies has been growing since 1990, with prevalence of 1343.85 thousand in 2019 [2]. The age-standardized death rate (ASDR) of 4.26, 1.42, 3.19, and 0.34 per 100,000 population in 2019, were respectively reported for leukemia, multiple myeloma, non-Hodgkin lymphoma, and Hodgkin lymphoma [3].

Blood infection (BSI) is a common and serious complication with a prevalence of 12- 46.2% in patients undergoing hematopoietic stem cell transplantation (HSCT) [4]. This high rate of blood infection is associated with delay in chemotherapy, delay in patient discharge, increase in treatment, and antimicrobial costs [5]. In recent years, the use of targeted molecular therapeutic methods has reduced the risk of neutropenia in this group of patients [6]. Despite these measures, bacterial blood infections are still considered an important cause of death in stem cell transplant recipients [7].

Both Gram-negative and Gram-positive microorganisms contribute to bloodstream infections of hematopoietic stem cell transplantation [8]. According to the geographical region, the type of microorganism and the pattern of microbial resistance is different. Today, due to the widespread use of broad-spectrum antibiotics, the rate of antibiotic resistance in strains of the Enterobacteriaceae family is increasing, of which a high quantity is attributed to Escherichia coli, whose prevalence in nosocomial infections is statistically significant [9,10,11].

The increasing prevalence of multidrug-resistant (MDR) strains has become a widespread global concern today [12]. This is because the treatment of infections caused by these strains is very difficult, especially in HSCT patients who have a weak immune system, and has resulted in a high mortality rate. Therefore, it is very important to address this issue among the clinicians. Therefore, due to the lack of a comprehensive study, we tried to investigate the prevalence of E. coli and antibiotic resistance patterns in this systematic review.

Methods

Search strategy

We conducted a systematic review of epidemiology of Escherichia coli in bloodstream infections of hematopoietic stem cell transplantation patients. For a comprehensive search of studies that reported the prevalence of E. coliand antibiotic resistance in bloodstream infections of patients undergoing HSCT from 2000 to January 1, 2024, databases such as PubMed, EMBASE, Google Scholar, Scopus, and Web of Science were searched. The main keywords used were: Escherichia coli, E. coli, epidemiology, bloodstream infection, BSI, microbial resistance, antibiotic resistance, hematopoietic malignancy, hematopoietic stem cell transplantation, and HSCT. The review was performed by two reviewers independently. If there was a disagreement about inclusion of studies between the researchers, it was resolved through discussion, otherwise, a third researcher was asked for advice.

Eligibility criteria

Studies that correctly reported epidemiology (prevalence and microbial resistance) were included in this study. Review articles (narrative reviews, systematic reviews, and meta-analyses), case reports, meetings, and conferences were excluded.

Evaluation of the quality

The evaluation of the quality of the studies was done by the Critical Appraisal Skills Programme (CASP) checklist (www.casp-uk.net), [13]. in which questions about the content of the studies, methodology, and how to present the results were raised. Finally, qualified studies were included and the remaining were excluded.

Extraction of data

Information such as first author, study type, geographical location, and year of publication, patients, bloodstream infection, and age were extracted and entered in designed forms.

Results

Study selection and characteristics

After searching, deleting duplicates, and applying all screening steps, finally, 30 studies were included in the present systematic review. Studies number from Turkey, Italy, Spain, Japan, Pakistan, Denmark, Thailand, Colombia, Germany, Brazil, China, Lebanon, Bulgaria, Belarus, Australia, USA, Egypt, South Korea, and other countries were respectively 2, 3, 2, 1, 1, 1, 1, 1, 1, 1, 3, 2, 1, 1, 1, 3, 1, 2, 1 (

Table 1. Features of articles included in the current review.

and Figure 1).

Prevalence of ESBL strains and mortality caused by E. coli strains

Prevalence of ESBLs strains of E. coli in bloodstream infections varied between 15-80%, whereas the mortality rate caused by E. coli strains in bloodstream infection varied between 6.7-27.3% (

Table 2. Features of articles listed in this review.

).

Antibiotic susceptibility in E. coli strains

As shown in Supplementary Table 1, resistance to ciprofloxacin, cefepime, third- and fourth-generation cephalosporins was reported to be the highest (prevalence of 100%), and the lowest was against amikacin with a prevalence of 13-38%. Also, resistance against fluoroquinolones was between 53-92%.

Discussion

As stated earlier, a large share of problems and deaths following bacterial blood infection in people undergoing stem cell transplantation is caused by Gram-negative bacteria, among which a significant part is represented by E. coli. For that, in our review, the prevalence of bacterial bloodstream infection amongst different studies varied between 8-41%. Also, BSI caused by E. coli varied between 2.5-57%. In some studies, a high BSI rate has been reported, for example, 51% by Macesic et al., [14]. while others have reported a moderate rate, and in some other studies, a low prevalence between 8% and 15% has been reported [15,16].

The findings of our review showed that there was no statistically significant difference between the prevalence of bacterial BSI and E.coli-related BSI in HSCT patients in different parts of the world, and in some cases, the prevalence rate in Asian countries was reported to be lower than in European and American countries. This could be because the bacteria identification techniques in Asian countries may be weak and unable to detect. Maybe it is because today, with the development and improvement in molecular diagnostic techniques, different countries have become closer to each other in microbial diagnosis. However, the existence of differences in reports can be related to the site of infection, clinical samples, operator precision, diagnostic phenotypic methods, type of molecular techniques used, and geographical region.

Findings of the current review indicated the BSI caused by E. colibetween 15-80% [3,4]. The high rate of prevalence of ESBL strains indicates an alarming increase in these strains, which must be taken seriously. This high trend is in accordance with the data obtained from other studies [3,17]. In addition, the data extracted from other studies showed that mortality caused by ESBL strains is higher compared to that cause by sensitive strains, which indicates a need for immediate intervention and targeted treatment to prevent mortality in HSCT patients, who are more susceptible to BSI caused by MDR strains [1,5,10,18,19].

The bloodstream infections caused by ESBLs became more important because studies from Asian and European countries reported a high prevalence.

Based on the results obtained from this review, resistance to ciprofloxacin, cefepime, and third- and fourth-generation cephalosporins was reported to be the highest, with a prevalence of 100%, and the lowest was against amikacin, with a prevalence of 13-38%. Also, resistance against fluoroquinolones was between 53-92%. These results indicated a very high resistance to antibiotics such as third generation cephalosporins and fluoroquinolones, and as a result, limit the choice of antibiotics for the treatment of BSI in patients undergoing HSCT, because American Society of Clinical Oncology/Infectious Diseases Society of America (ASCO/IDSA) recommends the use of fluoroquinolones for adult patients with immunosuppression associated with cancer at high risk of febrile neutropenia [20]. In addition, it should be noted that in hematological malignancies, the use of broad-spectrum antibiotic prophylaxis against Gram-negative bacteria, including E. coli, led to subsequent BSI with multidrug-resistant and ESBL pathogens [3].

As presented, a high resistance against all antibiotics used in these studies was reported, and the interesting thing is that only a study conducted by M. Yemişen et al. reported a low resistance to all antibiotics used [21]. High resistance to fluoroquinolones has been reported by Michele Malagola et al., [9] and Davide Mattei et al., [2] which are used as prophylaxis in patients undergoing HSCT to prevent bacteremia. Resistance to carbapenems has also been reported to be high, as Moghnieh et al., [22] and Wang et al. [23] reported a resistance between 75-83%, while Zeng et al. reported a resistance of about 16% [15]. This level of resistance is alarming; infection with carbapenemase-producing Enterobacteriaceae (KPC) strains has been reported to be 2%, while infection-related mortality has been reported to be 64% [9]. The difference in the level of antibiotic resistance can be due to reasons such as: the antibiogram method, geographical region, type of clinical sample, the amount of antibiotic use in each region, health and control measures against the spread of resistant strains, the amount of antibiotic use in agriculture and animal husbandry, and the antibiotic pressure due to the use of broad-spectrum antibiotics in each region [7].

It is important to investigate BSI caused by E. coli because the studies conducted in recent decades showed that the trend of infection has changed from Gram-positive microorganisms to Gram-negative bacteria, including E. coli [12,24]. Inappropriate empirical therapy can be one of the reasons for drug resistance, which has subsequently led to an increase in BSI with MDR and ESBL strains [4], where a high mortality rate due to E. coli-related BSI, as reported by several studies, prove this claim [23].

The most important limitation of this study was the lack of access to some articles, for which we contacted the authors but received no response. Another limitation was that languages other than English were not used for searching.

Conclusions

There is a high prevalence of E. coli-related BSI and subsequent mortality, especially by MDR and ESBL strains in patients undergoing HSCT, due to their greater susceptibility to infection. Essential measures are required to prevent the spread of microbial resistance. These measures can include monitoring colonization of patients, hand washing, avoiding arbitrary prescription of antibiotics, room isolation, personal protective equipment, proper use of disinfectants, non-use of broad-spectrum antibiotics, and prompt empirical antibiotic treatment.

Considering the challenge of BSI caused by MDR strains including E. coli in this group of patients, the use of alternative compounds such as plant extracts and nanoparticles should be considered.