Addressing the Burden of Antimicrobial Resistance in Vietnamese Hospitals

Vu Quoc, Dat (2020). Addressing the Burden of Antimicrobial Resistance in Vietnamese Hospitals. PhD thesis The Open University.



Hospital acquired infections (HAIs), especially ventilator associated respiratory infection (VARI) cause significant morbidity and mortality, and disproportionally so in low- and middle-income countries (LMICs), including Vietnam, where infection control in hospitals is often neglected. The management of HAIs in these settings is challenging because of the high proportions of antimicrobial drug resistance and limitations of laboratory diagnostics, financial and human resources in terms of knowledge and skills for antimicrobial stewardship and infection prevention and control.
Because resistance is driven by use of antimicrobials, my thesis started with a question on use and cost of antimicrobials in public hospitals in the country followed by a detailed assessment of use and cost of antimicrobials in the management of ventilator associated respiratory infections (VARI). I obtained detailed bids from hospitals and provincial departments of health representing 28.7% (1.68 / 5.85 billion US$) of the total hospital medication budget in Vietnam. Antimicrobials represented 28.6% of these costs. Antimicrobials were stratified using the Access, Watch, Reserve (AWaRe) groups proposed by WHO in 2017. I showed that the most commonly used antimicrobials across sites were second generation cephalosporins (20.3% of total procured defined daily dose, DDD) followed by combinations of penicillins and beta-lactamase inhibitors (18.4% of total procured DDD). The most expensive antimicrobials are the last resort antimicrobials, which can considerably increase the cost of treatment for patients with HAIs caused by multidrug resistant pathogens in critical care units in Vietnam. In recognition of this problem, I estimated the excess cost of management of VARI using a costing model study. At the current incidence rate of 21.7 episodes per 1000 ventilation-days, I estimated there were 34,428 episodes of VARI nationally, associated with a direct cost of more than US$ 40 million per year. Our studies showed the need for an affordable and scalable intervention in critical care units to reduce the burden of VARI and provide cost savings for national health expenditure.
My studies also showed that antimicrobial costs are a major component of the excess cost of VARI management in Vietnam (51.1%) and that a one day reduction in the duration of antimicrobial therapy can save US$ 1.72 million. Therefore, my thesis has focused on interventions to prevent VARI and to shorten antimicrobial therapy. In recognition of human resources constraints in Vietnam, including for microbiology diagnostics and critical care nursing, I have studied automatic technology and equipment, including matrix assisted laser desorption ionization-time of flight mass spectrometry (MALDITOF-MS) for rapid identification of pathogens and continuous automatic cuff pressure control device to prevent VARI. To examine effectiveness of these intervention, I conducted 2 randomised controlled trials to evaluate the clinical effectiveness of matrix assisted laser desorption ionization-time of flight mass spectrometry (MALDITOF-MS) in optimizing antimicrobial therapy and to evaluate the effectiveness of continuous cuff pressure control in preventing VARI. For the latter, pending unblinding and final results I describe the implementation of the trial and report the incidence of hospital acquired bloodstream infection during this trial. A diagnostic randomised controlled trial (RCT) was conducted to evaluate the impact of MALDITOF-MS versus conventional diagnostics in improving antimicrobial use in patients with confirmed infection. Although MALDITOF-MS provided more rapid identification of invasive bacterial and fungal pathogens than conventional microbiology, the proportion of patients on optimal therapy at 24 or 48 hours after growth of specimen did not increase. These findings showed that without human resources and an effective antimicrobial stewardship programme, technology alone cannot provide a solution for antimicrobial overuse in hospitals in LMICs. A randomized controlled clinical trial was conducted to evaluate the effectiveness of continuous cuff pressure control versus daily manual cuff measurement (VARI-prevent). In this study I recruited and followed-up 597 adult patients who were admitted to ICUs and were intubated within 48 hours of admission. The patients were randomised to receive either continuous or manual cuff pressure measurement and control and were followed for occurrence of VARI during ICU stay and up to 90 days after randomisation. The study has completed recruitment and follow-up and final analysis is ongoing. The overall rate of VARI and VAP in eligible patients was 23.7% (140/591) and 17.3% (102/591) respectively. The data from this trial (VARI-prevent) was analysed to estimate the incidence density rate of hospital acquired bloodstream infection (HABSI) in 3 ICUs in Vietnam for the first time. The most common pathogens causing HABSI were Klebsiella pneumoniae followed by Pseudomonas aeruginosa, Acinetobacter baumannii and Coagulase-Negative staphylococci. Polymicrobial culture results were reported in 6.8% (3/44) patients with culture confirmed HABSI. The rate of HABSI and central line associated BSI (CLABSI) were 7.4% (44/591) and 9.3% (31/333), respectively. The incidence density rate of HABSI and CLABSI were 3.76 per 1000 patients-days and 8.43 per 1000 catheter-days, respectively. This suggests that the implementation of infection prevention and control bundle including catheter care is important to reduce the high incidence of HABSI in Vietnam. The findings in my thesis are relevant to healthcare professionals and policy stakeholders. It demonstrates the magnitude of HAI burden and creates awareness of potential beneficial interventions. Results of my trials will be helpful to inform decisions to establish the antimicrobial stewardship programmes and infection prevention and control bundles to improve patients’ outcomes. [math mode missing closing $]

Viewing alternatives

Download history


Public Attention

Altmetrics from Altmetric

Number of Citations

Citations from Dimensions

Item Actions



  • Item ORO ID
  • 70201
  • Item Type
  • PhD Thesis
  • Academic Unit or School
  • Other Departments > Other Departments
  • Associated Research Centre
  • Oxford University Clinical Research Unit Vietnam
  • Copyright Holders
  • © 2020 The Author
  • Depositing User
  • Christopher Biggs