Enterobacter sakazakii can cause rare but life-threatening diseases such as meningitis in infants and neonates. Fourier transform infrared (FT-IR) spectroscopy was used to detect and discriminate between eight E. sakazakii strains, two Enterobacter cloacae strains, three Escherichia coli strains and two Klebsiella pneumoniae strains. FT-IR vibrational combination bands reflect subtle compositional differences in the cell membranes of E. sakazakii strains, especially in the region between 1,200 and 900 cm−1which contains absorption bands from carbohydrates. Two multivariate statistical analyses including principal component analysis (PCA) and soft independent modeling of class analogy (SIMCA) were used for data analysis. E. sakazakii strains were clearly distinguishable from the other strains by PCA. Based upon SIMCA analysis, 90% of E. sakazakii, 88% of E. cloacae, 91% of E. coli and 91% of K. pneumoniae samples were correctly classified, suggesting that this technique could be used to detect E. sakazakii strains rapidly and accurately.