Pulsed-field gel electrophoresis (PFGE) is the most common genotyping method used for the typing of a number of bacterial species. Generally, investigators use their own custom-developed protocol, but a standardized PFGE protocol would allow the comparison of typing results between laboratories and the tracing of strains around the country. In the present study, we optimized a PFGE protocol for subtyping of Acinetobacter baumannii, Escherichia coli and Klebsiella spp., which are commonly isolated from nosocomial infections in many hospitals. Reproducibility of our PFGE procedure was studied three times at 2- to 3-week intervals. Epidemiological concordance of the optimized PFGE procedure was tested on seven isolates of A. baumannii from a previous outbreak and seven A. baumannii isolates randomly selected among the clinical isolates. The optimized PFGE procedure was evaluated on a total of 174 clinical isolates including 62 A. baumannii, 50 E. coli and 62 Klebsiella spp. The inter-laboratory reproducibility of the optimized protocol was tested at four laboratories. The optimized procedure is completed in 28 h after culturing. It is likely to be cost-effective, due to the reduction in the time, reagent volume and enzyme concentration needed. The procedure showed high concordance with epidemiological data. There were no non-typeable isolates among the tested bacteria. It is reproducible and versatile. This protocol can be used to identify outbreaks and monitor the spreading rate of nosocomial infections caused by the tested bacterial isolates. Furthermore, due to its high intra- and inter-laboratory reproducibility, the protocol has the potential to be useful for comparing PFGE fingerprinting profiles of the isolates from different settings.