OBJECTIVES: Detection of carbapenem-hydrolysing class D β-lactamase (CHDL)-producing Acinetobacter spp. is critical for understanding antibiotic resistance. In this study, we compared the available detection techniques derived from the carbapenem inactivation method (CIM), using CHDL-producing Acinetobacter spp., and developed a modified method that uses bacterial lysate (lysate CIM; LCIM). METHODS: A total of 159 Acinetobacter spp. (102 carbapenemase producers and 57 non-producers) and 14 Pseudomonas spp. (7 carbapenemase producers and 7 non-producers) were tested. Modified CIM, simplified CIM, CIMTris, Triton-CIM and LCIM were compared using these strains. Distinct from the CIM, LCIM includes a longer incubation period (4 h) with 2.0% Triton X-100 (v/v) in 20 mM MOPS buffer instead of water. RESULTS: The sensitivity/specificity of the modified CIM, simplified CIM, CIMTris, Triton-CIM and LCIM were 71.6%/100%, 66.1%/89.1%, 88.1%/95.3%, 80.7%/100% and 97.2%/100%, respectively. LCIM was the most sensitive and specific. CONCLUSIONS: Use of bacterial lysate and MOPS increased the sensitivity of the CIM in detecting CHDL-producing Acinetobacter spp.