Statistical assessment of a laboratory method for growing biofilms.

Microbial biofilms have been grown in laboratories using a variety of different approaches. A laboratory biofilm reactor system, called the CDC biofilm reactor (CBR) system, has been devised for growing biofilms under moderate to high fluid shear stress. The reactor incorporates 24 removable biofilm growth surfaces (coupons) for sampling and analysing the biofilm. Following preliminary experiments to verify the utility of the CBR system for growing biofilms of several clinically relevant organisms, a standard operating procedure for growing a Pseudomonas aeruginosa biofilm was created. This paper presents the results of a rigorous, intra-laboratory, statistical evaluation of the repeatability and ruggedness of that procedure as well as the results of the experiments with clinically relevant organisms. For the statistical evaluations, the outcome of interest was the density (c.f.u. cm(-2)) of viable P. aeruginosa. Replicate experiments were conducted to assess the repeatability of the log density outcome. The mean P. aeruginosa log10 density was 7.1, independent of the coupon position within the reactor. The repeatability standard deviation of the log density based on one coupon per experiment was 0.59. Analysis of variance showed that the variability of the log density was 53 % attributable to within-experiment sources and 47 % attributable to between-experiments sources. The ruggedness evaluation applied response-surface design and regression analysis techniques, similar to those often used for sensitivity analyses in other fields of science and engineering. This approach provided a quantitative description of ruggedness; specifically, the amount the log density was altered by small adjustments to four key operational factors--time allowed for initial surface colonization, temperature, nutrient concentration, and fluid shear stress on the biofilm. The small size of the regression coefficient associated with each operational factor showed that the method was rugged; that is, relatively insensitive to minor perturbations of the four factors. These results demonstrate that the CBR system is a reliable experimental tool for growing a standard biofilm in the laboratory and that it can be adapted to study several different micro-organisms.