The objective of this study was to develop an optimized assay for Salmonella Typhi biofilm that mimics the environment of the gallbladder as an experimental model for chronic typhoid fever. Multi-factorial assays are difficult to optimize using traditional one-factor-at-a-time optimization methods. Response surface methodology (RSM) was used to optimize six key variables involved in S. Typhi biofilm formation on cholesterol-coated polypropylene 96-well microtiter plates. The results showed that bile (1.22%), glucose (2%), cholesterol (0.05%) and potassium chloride (0.25%) were critical factors affecting the amount of biofilm produced, but agitation (275 rpm) and sodium chloride (0.5%) had antagonistic effects on each other. Under these optimum conditions the maximum OD reading for biofilm formation was 3.4 (λ600 nm), and the coefficients of variation for intra-plate and inter-plate assays were 3% (n = 20) and 5% (n = 8), respectively. These results showed that RSM is an effective approach for biofilm assay optimization.
Twenty-eight (28) strains of ceftazidime-resistant Klebsiella pneumoniae were isolated from blood cultures of in-patients from University Hospital, Kuala Lumpur between March 1995 and August 1996. Three methods were used to detect the production of ESBL enzymes by these strains. These three methods include the double-disc synergy test (DDST), inhibitor-potentiated disc-diffusion test (IPDD) and the E-test ESBL method. All strains could be identified as ESBL producers using the DDST method by a minimum of two beta-lactams and these included either a combination of ceftazidime and ceftriaxone with clavulanate respectively or cefotaxime and aztreonam with clavulanate respectively. Similarly using a combination of either cefotaxime and ceftriaxone with clavulanate or ceftriaxone and aztreonam with clavulanate respectively would have detected all strains as ESBL producers. The IPDD method could also detect for ESBL activity based on combinations of beta-lactam antibiotics with clavulanate respectively. All combinations of beta-lactam antibiotics could detect for ESBL activity in all the strains except a combination of either ceftazidime and aztreonam or cefotaxime and ceftriaxone with clavulanate respectively. The E-Test method using ceftazidime alone and in combination with clavulanate was found to be the most effective method in the presumptive identification of ESBL activity in all the strains.