Monash University

Embargoed and Restricted Access

Reason: Under embargo until Feb 2019. After this date a copy can be supplied under Section 51(2) of the Australian Copyright Act 1968 by submitting a document delivery request through your library

Towards optimising the management of infectious diseases in high risk patients

posted on 2017-02-21, 00:04 authored by Doan, Tan Nhut
Hospital-acquired infections (HAIs) are a major cause of poor clinical outcomes and increased healthcare expenditure. In particular, patients receiving treatment for acute lymphoblastic leukaemia (ALL) and those in intensive care units (ICUs) are at high risk of HAIs. A significant proportion of HAIs can be prevented with optimal antimicrobial prophylaxis therapy, i.e. correct dose, dose interval and duration of therapy, and optimal infection control strategies, i.e. hand washing, patient isolation and ward cleaning. To date the approach to treat and prevent HAIs in patients with ALL and ICU patients, respectively, remains sub-optimal. This thesis provides important data to optimise the treatment and prevention of HAIs in these high risk patient groups. Specifically, this thesis will address the knowledge gaps relating to (i) the use of primary antifungal prophylaxis (PAP) in patients with ALL, (ii) the optimal dosing regimen for anidulafungin prophylaxis in patients with ALL, (iii) the transmission dynamics of Acinetobacter baumannii in ICUs and, (iv) the impact of infection control and antimicrobial stewardship interventions to prevent A. baumannii transmission in ICUs. As patients receiving chemotherapy for ALL are at an increased risk of invasive fungal disease (IFD), PAP should be considered. However, the optimal antifungal agent(s), dose and duration of PAP in patients with ALL are currently unknown. Consequently, it is essential to generate better information on how PAP is prescribed, the clinical outcomes and the economic consequences in this patient group to help optimise antifungal therapy. A retrospective multi-centre audit of patients with ALL undergoing chemotherapy was performed. PAP was only administered to 83/98 (85%) of participants. Importantly, PAP effectively reduced the incidence of proven/probable IFD by 19 percentage points compared with patients not receiving PAP (2.6% versus 21.4%, P = 0.024). Proven/probable IFD incurred substantial economic burden, increasing the median cost (hospitalisation, outpatient visits, IFD diagnostic tests, diagnostic procedures, antifungal drugs and surgical procedures for IFD treatment) by AU$121,520 (95% confidence interval AU$90,781-$180,141, P < 0.001) compared with those without IFD. For every five patients receiving PAP, one proven/probable IFD would be prevented, potentially resulting in a cost-saving of AU$121,520 for preventing one IFD case. To date, this is the largest multi-centre study on PAP use in patients with ALL, which suggests that PAP reduces IFD risk in these patients and is potentially cost-saving. Anidulafungin as a broad spectrum antifungal with a favourable pharmacokinetic (PK) and safety profile appears to be the ideal agent for PAP in patients with ALL. To date, dosing regimens for prophylactic use of anidulafungin have not been evaluated nor optimised in patients with ALL. Using a published two-compartment population PK model for anidulafungin, a Monte Carlo simulation platform in Berkeley Madonna was implemented to simulate the likely concentration-time profiles from five clinically relevant dosing regimens [200 mg loading dose (LD) on Day 1 then 100 mg daily, 200 mg LD on Day 1 then 100 mg every 48 hours (q48 h), 200 mg q48 h, 200 mg q72 h, 300 mg q72 h]. For each dosing regimen, the probability of target attainment (PTA) against Candida spp. was calculated. The pharmacodynamic targets for Candida spp. were a free-drug area under the plasma concentration-time curve over the minimum inhibitory concentration ratio (ƒAUC/MIC) ≥ 13 and a free-drug maximum plasma concentration over MIC ratio (ƒCmax/MIC) ≥ 0.7. The currently recommended dosing regimen, i.e. 200 mg LD on Day 1 then 100 mg daily, was not optimal (PTA < 90%). Intermittent dosing regimens, i.e. high doses combined with prolonged dosing intervals, achieved better PTA. This study is the first to optimise anidulafungin therapy using Monte Carlo simulations. These findings support the potential use of intermittent dosing of anidulafungin for PAP in patients being treated for ALL. Further studies evaluating the clinical and economic outcomes are required to support the proposed dosing regimens. The transmission of nosocomial pathogens is a key driver for HAIs. Infections caused by A. baumannii are difficult to treat because it is resistant to most antibiotics and standard hospital cleaning procedures. Accordingly, an in-depth understanding of the transmission dynamics of A. baumannii is crucial for its containment within the hospital. Because the underlying transmission process is not fully observable, characterising the transmission dynamics of A. baumannii can be challenging. In addition, most data recorded in hospitals and healthcare systems only record the time of symptom onset, and therefore the exact time of acquisition is normally unknown. A hidden Markov model (HMM) with Bayesian inference was utilised to overcome these challenges and to estimate the unknown transmission parameters, i.e. cross-transmission coefficient and sporadic acquisition coefficient. Longitudinal data set was collated from previous studies describing the number of patients colonised with A. baumannii in the ICU of three hospitals. Utilising the HMM and the longitudinal data set, the expected time to A. baumannii colonisation as a result of cross-transmission for each susceptible patient in Hospital 1 and 3 was estimated to be 20 and 31 days, respectively, while there was little cross-transmission in Hospital 2. The expected time for one new colonised case to arise from sporadic acquisition in the three hospitals was 39, 17 and 109 days, respectively. The basic reproduction ratio, R0, for Hospital 1, 2 and 3 was 1.5, 0.02 and 1.6, respectively. This is the first time that a transmission dynamic model has been used to characterise the transmission of A. baumannii, providing important insight into the spread of this pathogen. The aforementioned study clearly identifies the need to optimise infection control to prevent A. baumannii acquisition in the ICU. To implement optimal infection controlstrategies for A. baumannii in hospitals, the impact of interventions against this pathogen must be evaluated and understood. Accordingly, a comprehensive and biologically plausible transmission dynamic model was developed to quantify the effects of various interventions on A. baumannii transmission for a 100-bed ICU setting. The novel aspects of this study are the implications on transmission of the environmental reservoir and the intestinal microbiota. The model predicted that improving compliance with hand hygiene to ≥ 87%, daily ward cleaning, reducing the length of stay of colonised patients to ≤ 10 days, or reducing the durations of antibiotic treatment to ≤ 6 days are independent factors that can be optimised to prevent A. baumannii acquisition in the ICU, reducing R0 to < 1. This thesis has provided an improved understanding of PAP in patients with ALL and, the transmission characteristics of A. baumannii in the ICU. It provides the first quantitative data in support of PAP use in ALL patients. The findings also provide a rationale for further clinical investigations of intermittent dosing of anidulafungin prophylaxis in the ALL setting. The data generated will assist in optimising infection control strategies for A. baumannii infections in ICUs.


Campus location


Principal supervisor

David Kong

Additional supervisor 1

Carl Kirkpatrick

Additional supervisor 2

Emma McBryde

Year of Award


Department, School or Centre

Centre for Medicine Use and Safety (CMUS)


Doctor of Philosophy

Degree Type



Faculty of Pharmacy and Pharmaceutical Sciences

Usage metrics

    Faculty of Pharmacy and Pharmaceutical Sciences Theses


    No categories selected


    Ref. manager