Multiplex droplet digital polymerase chain reaction for rapid diagnosing suspected bloodstream infections in patients with hematologic malignancies

Introduction

Hematological disorders, particularly malignant ones, often necessitate high-dose chemotherapy, rendering patients immunocompromised and highly susceptible to various microbial infections (1,2). Infections left untreated in these patients can lead to prolonged treatment durations and increased healthcare costs, with severe cases posing a threat to patient lives. The 2021 International Guidelines for Management of Sepsis and Septic Shock recommend the administration of antimicrobial therapy as soon as possible, ideally within one hour of diagnosis (3). Fever with neutropenia is commonly observed in hematological disease patients, and the gold standard for pathogen identification and antimicrobial susceptibility testing is blood culture (BC). However, BCs suffer from low positivity rates (less than 10% to about 50%) (4-6), inability to confirm true infections or identify pathogen types, and extended processing time making it fail to meet the recommendations of timely antimicrobial therapy. Additionally, the protracted turnaround time and low positivity rate of BCs may delay the implementation of chemotherapy regimens and potentially lead to antibiotic overuse (7). There is a pressing clinical need for a rapid and accurate diagnostic method capable of identifying infectious pathogens and drug-resistant genes.

To address the limitations of BCs, numerous technologies are currently applied for the detection of bloodstream infections (BSIs), such as real-time quantitative polymerase chain reaction (qPCR) (8), next generation sequencing (NGS) (9), microarray-based assay (10), and microfluidics-based assays (11). However, the sensitivity of these technologies falls short of clinical requirements, or their costs are prohibitively high, rendering them temporarily impractical for clinical use (9). In recent years, droplet digital polymerase chain reaction (ddPCR) has demonstrated increasing clinical value in the detection of pathogen infections (2,12,13). ddPCR allows for the absolute quantification of nucleic acids without the need for standard curves or reference samples. It possesses a high degree of accuracy and can precisely detect single-copy nucleic acids, thus exhibiting outstanding sensitivity. This technology has extensive applications in various fields, including the monitoring of minimal residual disease in cancer, non-invasive prenatal testing for genetic abnormalities, and the surveillance of molecular mutations, among others (14-18). Compared to conventional BC, digital polymerase chain reaction (PCR) can reduce the time required for infection reporting from several days to approximately 4 hours, representing an innovation in diagnostic technology. ddPCR partitions samples into thousands of droplets for absolute nucleic acid quantification without a standard curve, offering higher sensitivity, precision, and tolerance to inhibitors than conventional PCR. Digital PCR has significant applications in the management of infections in hematologic malignancy patients. Its high sensitivity, speed, and precision facilitate rapid diagnosis of infectious pathogens and drug-resistant genes. This enables physicians to implement targeted treatments promptly, enhancing treatment success rates. Additionally, digital PCR can monitor changes in pathogen types and copy numbers, guiding adjustments in treatment strategies. The quick reporting time associated with this technology reduces treatment delays, lowers the risk to patients’ lives, and conserves healthcare resources.

This study aimed to utilize a digital PCR-based platform for detecting pathogenic nucleic acids and drug-resistant genes in blood, as a solution for BSI detection. It seeks to assess and validate the feasibility of applying this approach for pathogen and drug-resistant gene detection in hematologic malignancy patients. The study intended to provide theoretical support for further evaluating the feasibility and clinical value of this BSI detection method in guiding antibiotic therapy. We present this article in accordance with the MDAR reporting checklist (available at https://tcr.amegroups.com/article/view/10.21037/tcr-23-2240/rc).

Methods

Study population

This investigation involved a prospective pilot diagnostic study aiming at clinically validating the effectiveness of the multiplex ddPCR panel for diagnosing suspected BSIs in febrile neutropenia in patients with hematologic malignancies who are currently undergoing or have recently undergone chemotherapy or immunotherapy. The research was conducted at the hospital for hematology intensive care unit (HCU) in Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, spanning from August 2022 to November 2022. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of Medicine. Informed consent was obtained from each included participant.

The sample size was determined by the number of cases in the area over the study period. A total of 47 patients were included in our study. Inclusion criteria encompassed inpatients diagnosed with acute myeloid leukemia (AML), acute lymphoblastic leukemia (ALL), or high-risk myelodysplastic syndrome (MDS), experiencing febrile neutropenia. The risk stratification of the hematological malignancies follows the Chinese Society of Clinical Oncology (CSCO) guidelines for the Diagnosis and Treatment of Malignant Hematological Diseases 2023. Exclusion criteria encompassed patients in imminent life-threatening conditions or severe shock with unsuccessful resuscitation, individuals with significant liver or renal dysfunction, and those in the end-stage of disease.

Sample collection and BSI diagnosis

On day 1 (the first day of febrile neutropenia), simultaneous to routine clinical BC, >5 mL of patient blood was collected using specialized blood collection tubes for subsequent ddPCR testing. In cases where BC was drawn during non-working hours, a simultaneous sample was collected using specialized tubes (Thermo Scientific™ Matrix™ 2D Barcoded ScrewTop Storage Tubes, Thermo Fisher Scientific, Hudson, NH, USA), stored appropriately (4 ℃ or room temperature, avoiding freezing), and delivered to the laboratory within 72 hours. Bacterial, fungal, and resistant genes were concurrently detected after nucleic acid extraction. For patients with positive ddPCR results on day 1, a follow-up blood sample was collected on day 3 for bacterial, fungal, and resistant genes testing. This aimed to observe changes in pathogen types and copy numbers, assessing their correlation with treatment plans and clinical manifestations. If necessary, subsequent blood draws on day 5 allowed for continuous monitoring of changes in pathogen types and copy numbers. In cases where BCs were negative but ddPCR results were positive, a recommendation was made to repeat BC on day 3 or day 5 while concurrently conducting ddPCR testing. This approach helps resolve discrepancies and provides valuable insights into the infection dynamics. Aseptically inoculate blood into aerobic, anaerobic, and/or fungal-specific culture bottles. Incubate using an automated continuous monitoring system (BACTEC) at 35–37 ℃. Monitor for microbial growth for up to 5 days (bacteria) or 2 weeks (fungi), subculturing positive samples for identification.

Target microorganisms

Bacteria: the study focused on detecting the following bacterial pathogens—Pseudomonas aeruginosa, Escherichia coli, Klebsiella pneumoniae, Staphylococcus epidermidis, Staphylococcus aureus, Enterococcus faecalis, Staphylococcus haemolyticus, Acinetobacter baumannii, Staphylococcus lugdunensis, Enterobacter cloacae, Staphylococcus capitis, Enterococcus faecium, Streptococcus pneumoniae, Citrobacter freundii, Stenotrophomonas maltophilia. Fungi: detection was conducted for Candida albicans, Candida glabrata, Candida parapsilosis, and Candida tropicalis. Resistant genes: evaluation included screening for carbapenem-resistant genes (blaKPC, blaNDM) and the methicillin-resistant gene (mecA).

Plasma DNA extraction and ddPCR testing

Plasma DNA extraction and multiplex ddPCR testing involved a five-channel system capable of detecting the eight most prevalent bacterial pathogens, four fungal pathogens, and three antimicrobial resistance (AMR) genes directly from blood at concentrations as low as 50 copies/mL (Pilot Gene Technologies, Hangzhou, China). The detection depth of ddPCR, determined by testing whole blood specimens spiked with various concentrations of different microbial species, was found to be 50 copies/mL, except for blaKPC (80 copies/mL). Whole-blood samples were stored at 4 ℃, and ddPCR testing was conducted either on the same day or the following day, aligning with the suspected BSI timing. The ddPCR procedures, lasting approximately 2.5 hours, adhered to the manufacturer’s protocol. Samples underwent plasma conversion through centrifugation (1,600 r.c.f. for 15 minutes), requiring about 40 minutes for sample preparation. Subsequently, the reaction mixture in the sample cup traversed the micro-channel (Droplet Generator DG32, Hercules, CA, USA) under pressure, generating tens of thousands of water-in-oil emulsion droplets within 20 minutes due to gravity and shear force. After PCR amplification for 60 minutes using Thermal Cycler TC1, droplet count and amplitude scanning, as well as data analysis, were completed within 30 minutes using a chip scanner CS5 and GenePMS software (v2.0.01.20011). A synthesized DNA fragment served as the positive control, while DNase-free water or blood samples from three healthy subjects functioned as negative controls. The copies of each targeted pathogen or gene were reported through ddPCR results.

Definitions and clinical data

Culture-proven BSI is characterized by positive BCs in a patient exhibiting systemic signs of infection, and it may either be secondary to a documented source or primary, as per the definitions provided by the National Healthcare Safety Network (https://www.cdc.gov/nhsn/pdfs/pscmanual/4psc_clabscurrent). Patient clinical data, encompassing demographics, comorbidities, organ dysfunction, surgical interventions, and clinical outcomes, were extracted from medical records.

Two HCU physicians independently verified the results of ddPCRs and BC (BCs). Suspected infectious episodes from routine microbiological cultures, including BCs, abdominal, respiratory, skin and soft tissue, as well as other tissue/fluid cultures, were collected within 7 days of enrollment following standard microbiology laboratory procedures. A comprehensive clinical infection standard was defined, comprising all microbiological findings in conjunction with clinical adjudication.

Interpretation of BSI and ddPCR results

BSI results related to ddPCR-targeted pathogens or AMR genes were synthesized for further data scrutiny. A positive ddPCR result was determined if one or more target bacteria were detected, while it was deemed negative if no targets were identified. Polymicrobial infection was characterized by the detection of more than one microorganism, identified either through ddPCR or BC. BSI and ddPCR outcomes were categorized as either concordant (both positive and negative) or discordant. Instances where BCs were positive, but ddPCR results were negative or divergent, were designated as presumptive false-negative cases. To reconcile disparities, discordant ddPCR+/BSI− results were classified as probable BSI, possible BSI, or presumptive false-positive cases, aligning with methodologies from prior studies. Definitions for each classification were as follows: (I) probable—ddPCR result concordant with a microbiological test performed within seven days from an alternative site other than blood; (II) possible—lacking microbiological data, but ddPCR result showing potential pathogenicity based on clinical presentation and laboratory findings; (III) presumptive false-positive—ddPCR result inconsistent with clinical presentation.

Statistical analysis

Primary outcomes focused on assessing the sensitivity and specificity of ddPCR testing, determined by comparing positive BC results with ddPCR-targeted pathogens and AMR genes. Secondary outcomes involved the clinical validation of ddPCR testing for suspected BSIs, comparing results with all microbiological cultures and composite clinical diagnoses. Sensitivity, specificity, and positive and negative predictive values were computed based on these outcomes. Per-assay calculations treated individual pathogens in each sample separately.

Statistical analysis utilized IBM SPSS Statistics software (v23.0) (IBM, Armonk, NY, USA). Continuous variables were presented as the median and interquartile range (IQR), while categorical variables were reported as frequencies and percentages. The disparity in positivity rates between BCs and ddPCRs was assessed using the Chi-squared test. Statistical significance was attributed to differences where P values were ≤0.05.

Results

Patient clinical characteristics and pathogens detection results

A total of 47 patients with hematological malignancy containing 154 episodes (BC and ddPCR simultaneously) suspected as BSIs were consecutively included in the study. For BC tests, one pathogen infection was detected 5 times, and two pathogens concurrent infection happened once. Meanwhile, a broader spectrum of infectious time and space was demonstrated in ddPCR detections with a single pathogen identified 17 times and two pathogens identified 3 times (Figure 1). The characteristics of patients are listed in Table 1. Overall, five patients (10.64%) suffered from solid tumors, two patients (4.26%) suffered from immunosuppressive disease and four patients (8.51%) undergone circulation system disease.

Figure 1 The frequency distribution of pathogens detected by BC and ddPCR methods. (A) For BC tests, one pathogen infection was detected 5 times, and two pathogens concurrent infection happened once; (B) a broader spectrum of infections at the same time and space was demonstrated in ddPCR detections with single pathogens identified 17 times and two pathogens identified 3 times. BC, blood culture; ddPCR, droplet digital polymerase chain reaction.

Pathogens detected by BC were Gram-negative bacteria (n=3, Acinetobacter baumannii, Pseudomonas aeruginosa and Stenotrophomonas maltophilia) and Gram-positive bacteria (n=2, Streptococcus dysgalactiae and Staphylococcus epidermidis). Unfortunately, the Gram bacteria episodes of Gram-positive bacteria had no simultaneous ddPCR test results. For ddPCR set, 24 positive episodes were confirmed including Candida albicans (n=1), Enterococcus faecium (n=1), Candida parapsilosis (n=1), Staphylococcus capitis (n=2), Staphylococcus aureus (n=1), Staphylococcus hominins (n=1), Staphylococcus epidermidis (n=5), Pseudomonas aeruginosa (n=4), Stenotrophomonas maltophilia (n=2), Klebsiella pneumoniae (n=6). All of the pathogens detected in the present study are listed in Figure 2.

Figure 2 Distribution of pathogens and the number of times detected by ddPCR. ddPCR, droplet digital polymerase chain reaction.

Of all the 47 patients, the ones who suffered from AML accounted for 78.72% (n=37), the ones with ALL shared 12.77% (n=6), and the ones with MDS occupied 8.51% (n=4). Patients in the favorable group occupied the majority part (n=15) of the AML set. However, patients in the adverse group were all from the four BC positive patients and among them, one of the patients with TP53^mut endured two kinds of pathogens (including Staphylococcus epidermidis identified in the blood samples from both the left and right arms respectively, and Acinetobacter baumannii). The favorable group of patients shared only one case of ddPCR positive result (Candida albicans) and so did the intermediate group (Staphylococcus epidermidis). In addition, the BC results of the two groups were totally negative. It is worth pointing out that the patients with ALL, even those classified as in the adverse group, did not show positive ddPCR results, and the patients with MDS were free of TP53 mutation.

Overall comparison between BCs and ddPCRs

No concordantly positive were found in 154 episodes. BC and ddPCR yielded positive results for bacteria in five patients (10.64%) and 14 patients (29.79%), respectively. Two (50%) of all the four patients had a history of sarcoma and sicca syndrome, respectively. The one with sicca syndrome in combination with Stenotrophomonas maltophilia ranked among the two individuals who showed severe pulmonary computerized tomography (CT) lesions.

The 24 positive episodes (involving 14 patients) for ddPCR contained 10 episodes (41.67%) of Gram-positive bacteria, 12 episodes (50%) of Gram-negative bacteria, and two episodes (8.33%) of fungi. Patients from favorable, intermediate, and adverse group were six (42.86%), three (21.43%), and five (35.71%). The only positive BC episode that was capable of one-to-one correspondence with ddPCR results was Streptococcus dysgalactiae, but the simultaneous ddRCR result indicated the Stenotrophomonas maltophilia infection. In ddPCR positive patients, 35.71% (5/14) had pulmonary CT lesions. Among the 14 PCR-positive patients, four were found to have drug resistance genes, including one case of KPC (Staphylococcus aureus) and three cases of mecA (one Staphylococcus epidermidis and two Pseudomonas aeruginosa).

To summarize the discordant results between BC and ddPCR, we conducted the positive and negative comparison between them (Table 2). No BC+/ddPCR+ pathogens were confirmed while one episode of BC+/ddPCR− was identified. Seventeen coupled BC−/ddPCR+ pathogens were detected including seven Gram-positive bacteria, 10 Gram-positive bacteria and Candida albicans. Considering no episodes were compatible between the two detection methods, only specificity was calculated. The total specificity for ddPCR to BC was 81.44%. Taking the background as all the microbiological testing, the sensitivity was 15% and the specificity was 83.3% (Table 3). Among the 14 PCR-positive patients, four were found to have drug resistance genes, including one case of KPC and three cases of mecA.

Discussion

The threat of BSI in hematological malignancies cannot be underestimated. Swift identification of the pathogenic agent and subsequent rational selection of anti-infective therapy hold immense significance for this population: they ensure the maintenance of chemotherapy intensity and duration, thereby augmenting anti-tumor efficacy (19). This highlights the critical importance of timely and accurate diagnostic and therapeutic decisions in mitigating the risks associated with BSI, ultimately optimizing patient outcomes in the context of hematological malignancies.

BC yielded positive results for bacteria only in five patients (10.64%). Using the traditional pathogenic culture to detect bacteria required pre-enrichment, isolation and purification, biochemical identification, and serological identification, which would take 3–7 days. The traditional culture method is prone to give false-negative results (20).

The absence of immune function and the reduction of white blood cells in patients with hematological malignancies often lead to atypical manifestations of inflammation. Consequently, diagnostic methods such as pulmonary CT scans and sputum cultures frequently do not demonstrate typical results, creating difficulties in diagnosis and treatment. This further emphasizes the need for a sophisticated and comprehensive approach in managing the complex healthcare needs of these patients.

NGS is a powerful technology capable of detecting and analyzing complex biological samples. However, like any other technology, NGS also has its limitations. In the context of infection detection, these may include complexities in sample preparation and processing, challenges in data analysis, and the potential for false positive or false negative results (21). These limitations can potentially impact the accuracy and reliability of NGS in infection detection.

In this study, BC and ddPCR yielded positive results for bacteria only in five patients (10.64%) and 14 patients (29.79%), respectively, with ddPCR demonstrating higher positive rate (81.44%). Our findings indicated that multiplex ddPCR outperformed traditional BC methods in terms of positive rate and time effectiveness (22), however, whether it is a true positive rate requires further large-scale verification in the future. Multi-timepoint and multi-site synchronous sampling combined with dynamic load analysis constitutes an effective strategy to enhance the diagnostic specificity of ddPCR, significantly reducing overtreatment due to sample contamination. Future implementation requires cost-effectiveness analysis and automation technologies to facilitate clinical adoption.

The outperformance ensures that subsequent treatments are informed by more accurate and timely data, facilitating the selection of more effective anti-infective strategies. For hematological malignancy patients, this translates into swifter identification of pathogenic agents and initiation of targeted therapies, thereby mitigating the risks of infection-related complications and enhancing the efficacy of anti-cancer treatments. It is worth mentioning that the detection of skin-associated bacteria such as Staphylococcus capitis, S. epidermidis, and S. hominis and we have established a three-tier framework to differentiate contamination from true infection. Considering the patient’s fever symptoms, we still used antibiotics for positive bacteria when staphylococci were detected, and the patient’s treatment was effective.

Our research further reveals that multiplex ddPCR offers a broader detection range for pathogenic bacteria compared to BC methods. Regarding the breadth of detection, ddPCR identified 10 different pathogens, while only two pathogens went undetected. In contrast, BC detected only five different pathogens. Hematological malignancy patients, due to their disease and associated treatments, often experience immune deficiencies, which heighten their susceptibility to multiple infections. The extensive coverage of ddPCR allows for a more comprehensive understanding of a patient’s infection profile, informing personalized treatment strategies more effectively.

Among the 14 PCR-positive patients, four were found to have drug resistance genes (one case of KPC and three cases of mecA). Multiplex ddPCR also offers the distinct advantage of detecting resistance genes. Hematological malignancy patients, due to their immune deficiencies, often have higher rates of antibiotic usage in terms of probability, frequency, and intensity compared to the general population. This not only increases the incidence of drug resistance but also complicates the task of antibiotics stewardship (23). The ability of multiplex ddPCR to detect resistance genes empowers clinicians to make more informed and precise choices when selecting anti-infective treatments, thereby enhancing therapeutic outcomes and mitigating the risks of drug resistance emergence.

Despite the significant potential of multiplex digital PCR in diagnosing infections associated with hematological malignancies, we acknowledge that there are currently several limitations to this technology. These include issues with sensitivity and specificity, challenges in data analysis, cost and accessibility constraints, as well as the need for further standardization and optimization of the technique. These factors can impact the reliability and reproducibility of results, underscoring the importance of continued research and technological advancements in this field (24).

One limitation of our study is the small sample size, which may impact the broad applicability and reliability of our findings. Future studies are warranted to further validate our observations through the inclusion of a larger sample. Apart from the small sample size, there are other limitations to our study. We primarily focused on infection profiles in patients with AML, ALL or MDS, without considering patients with other subtypes of hematological malignancies. Moreover, our investigation centered on utilizing multiplex ddPCR technology for pathogenic bacteria detection without engaging in direct comparisons with other new diagnostic modalities such as NGS. In addition, in all cases, BC and ddPCR did not identify the same bacterial species. This discrepancy—likely reflecting the detection of viable, proliferating bacteria versus the detection of bacterial DNA (ddPCR)—should be explored further, particularly in terms of clinical relevance. Consequently, our findings necessitate further validation in broader populations and across multiple diagnostic approaches.

Conclusions

In summary, multiplex ddPCR is a flexible and versatile platform that can be used as a supplement to traditional BC. When combined with clinical evidence of infection, ddPCR shows potential advantages in rapidly diagnosing suspected BSIs and AMR genes in clinical practice for blood cancer treatment. With the continuous development and application of new technologies, we can expect to see further improvements and refinements in multiplex ddPCR technology in the future.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the MDAR reporting checklist. Available at https://tcr.amegroups.com/article/view/10.21037/tcr-23-2240/rc

Data Sharing Statement: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-23-2240/dss

Peer Review File: Available at https://tcr.amegroups.com/article/view/10.21037/tcr-23-2240/prf

Funding: This study was supported by Scientific Research Project of Shanghai Municipal Health Commission (No. 202040040), Zhejiang Provincial Science and Technology Plan Project (No. 2022C01007), National Natural Science Foundation Grants of China (No. 82200166), Natural Science Foundation of Tibet Autonomous Region [Nos. XZ2019ZR-ZY50(Z), XZ2022ZR-ZY31(Z)].

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tcr.amegroups.com/article/view/10.21037/tcr-23-2240/coif). All authors declared that this study was supported by Scientific Research Project of Shanghai Municipal Health Commission (No. 202040040), Zhejiang Provincial Science and Technology Plan Project (No. 2022C01007), National Natural Science Foundation Grants of China (No. 82200166), Natural Science Foundation of Tibet Autonomous Region [Nos. XZ2019ZR-ZY50(Z), XZ2022ZR-ZY31(Z)]. J.X. is an employee of Pilot Gene Technology (Hangzhou) Co., Ltd. The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of Medicine. Informed consent was obtained from each included participant.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Kanafani ZA, Dakdouki GK, El-Chammas KI, et al. Bloodstream infections in febrile neutropenic patients at a tertiary care center in Lebanon: a view of the past decade. Int J Infect Dis 2007;11:450-3. [Crossref] [PubMed]
2. McMahon S, Sahasrabhojane P, Kim J, et al. Contribution of the Oral and Gastrointestinal Microbiomes to Bloodstream Infections in Leukemia Patients. Microbiol Spectr 2023;11:e0041523. [Crossref] [PubMed]
3. Evans L, Rhodes A, Alhazzani W, et al. Surviving sepsis campaign: international guidelines for management of sepsis and septic shock 2021. Intensive Care Med 2021;47:1181-247. [Crossref] [PubMed]
4. Murray PR, Masur H. Current approaches to the diagnosis of bacterial and fungal bloodstream infections in the intensive care unit. Crit Care Med 2012;40:3277-82. [Crossref] [PubMed]
5. Pilecky M, Schildberger A, Orth-Höller D, et al. Pathogen enrichment from human whole blood for the diagnosis of bloodstream infection: Prospects and limitations. Diagn Microbiol Infect Dis 2019;94:7-14. [Crossref] [PubMed]
6. Opota O, Croxatto A, Prod'hom G, et al. Blood culture-based diagnosis of bacteraemia: state of the art. Clin Microbiol Infect 2015;21:313-22. [Crossref] [PubMed]
7. Fabre V, Carroll KC, Cosgrove SE. Blood Culture Utilization in the Hospital Setting: a Call for Diagnostic Stewardship. J Clin Microbiol 2022;60:e0100521. [Crossref] [PubMed]
8. van den Brand M, van den Dungen FAM, Bos MP, et al. Evaluation of a real-time PCR assay for detection and quantification of bacterial DNA directly in blood of preterm neonates with suspected late-onset sepsis. Crit Care 2018;22:105. [Crossref] [PubMed]
9. Hu B, Tao Y, Shao Z, et al. A Comparison of Blood Pathogen Detection Among Droplet Digital PCR, Metagenomic Next-Generation Sequencing, and Blood Culture in Critically Ill Patients With Suspected Bloodstream Infections. Front Microbiol 2021;12:641202. [Crossref] [PubMed]
10. Walker T, Dumadag S, Lee CJ, et al. Clinical Impact of Laboratory Implementation of Verigene BC-GN Microarray-Based Assay for Detection of Gram-Negative Bacteria in Positive Blood Cultures. J Clin Microbiol 2016;54:1789-96. [Crossref] [PubMed]
11. Zhang C, Zheng X, Zhao C, et al. Detection of pathogenic microorganisms from bloodstream infection specimens using TaqMan array card technology. Sci Rep 2018;8:12828. [Crossref] [PubMed]
12. Li M, Zhao L, Zhu Y, et al. Clinical value of droplet digital PCR in the diagnosis and dynamic monitoring of suspected bacterial bloodstream infections. Clin Chim Acta 2023;550:117566. [Crossref] [PubMed]
13. Lin K, Zhao Y, Xu B, et al. Clinical Diagnostic Performance of Droplet Digital PCR for Suspected Bloodstream Infections. Microbiol Spectr 2023;11:e0137822. [Crossref] [PubMed]
14. Wouters Y, Dalloyaux D, Christenhusz A, et al. Droplet digital polymerase chain reaction for rapid broad-spectrum detection of bloodstream infections. Microb Biotechnol 2020;13:657-68. [Crossref] [PubMed]
15. Oellerich M, Schütz E, Beck J, et al. Using circulating cell-free DNA to monitor personalized cancer therapy. Crit Rev Clin Lab Sci 2017;54:205-18. [Crossref] [PubMed]
16. Postel M, Roosen A, Laurent-Puig P, et al. Droplet-based digital PCR and next generation sequencing for monitoring circulating tumor DNA: a cancer diagnostic perspective. Expert Rev Mol Diagn 2018;18:7-17. [Crossref] [PubMed]
17. Tan C, Chen X, Wang F, et al. A multiplex droplet digital PCR assay for non-invasive prenatal testing of fetal aneuploidies. Analyst 2019;144:2239-47. [Crossref] [PubMed]
18. Galimberti S, Genuardi E, Mazziotta F, et al. The Minimal Residual Disease in Non-Hodgkin's Lymphomas: From the Laboratory to the Clinical Practice. Front Oncol 2019;9:528. [Crossref] [PubMed]
19. Wang J, Wang M, Zhao A, et al. Microbiology and prognostic prediction model of bloodstream infection in patients with hematological malignancies. Front Cell Infect Microbiol 2023;13:1167638. [Crossref] [PubMed]
20. Ahlstrand E, Bäckman A, Persson L, et al. Evaluation of a PCR method to determine the clinical significance of blood cultures with Staphylococcus epidermidis in patients with hematological malignancies. APMIS 2014;122:539-44. [Crossref] [PubMed]
21. Qin C, Zhang S, Zhao Y, et al. Diagnostic value of metagenomic next-generation sequencing in sepsis and bloodstream infection. Front Cell Infect Microbiol 2023;13:1117987. [Crossref] [PubMed]
22. Wu J, Tang B, Qiu Y, et al. Clinical validation of a multiplex droplet digital PCR for diagnosing suspected bloodstream infections in ICU practice: a promising diagnostic tool. Crit Care 2022;26:243. [Crossref] [PubMed]
23. El Sherif HM, Elsayed M, El-Ansary MR, et al. BioFire FilmArray BCID2 versus VITEK-2 System in Determining Microbial Etiology and Antibiotic-Resistant Genes of Pathogens Recovered from Central Line-Associated Bloodstream Infections. Biology (Basel) 2022;11:1573. [Crossref] [PubMed]
24. Liu W, Wang C, Pan F, et al. Clinical Application of a Multiplex Droplet Digital PCR in the Rapid Diagnosis of Children with Suspected Bloodstream Infections. Pathogens 2023;12:719. [Crossref] [PubMed]