Overprediction of Carbapenemase-Containing Isolates by an Automated AST Instrument's Computer Algorithm

Area of Science:

• Clinical Microbiology
• Infectious Diseases
• Antimicrobial Resistance

Background:

• The VITEK 2 system is widely used for predicting antibiotic resistance mechanisms.
• However, its reliability in detecting carbapenemases is questionable due to inconsistent research findings.
• Existing methods and results for carbapenemase prediction vary significantly.

Purpose of the Study:

• To evaluate the clinical reliability of VITEK 2 and Advanced Expert System (AES) for carbapenemase detection.
• To compare VITEK 2 and AES predictions against a modified carbapenemase inactivation method.
• To assess the utility of these systems in a routine academic medical center laboratory workflow.

Main Methods:

• A comparative study was conducted in an academic medical center laboratory.
• VITEK 2 and AES predictions were compared with the modified carbapenemase inactivation method.
• The study focused on routine workflow for carbapenemase detection in a specific patient population.

Main Results:

• The positive predictive value for carbapenemase detection using VITEK 2 and AES was found to be 30.3% (91 out of 300 cases).
• This indicates a significant rate of false positives in identifying carbapenemase-producing organisms.
• The findings highlight limitations in the accuracy of these automated systems for definitive carbapenemase identification.

Conclusions:

• The VITEK 2 and AES are not reliable as definitive tests for carbapenemase detection.
• These systems should be utilized as a preliminary screening tool for carbapenemase-containing organisms.
• Further investigation and confirmatory testing are recommended for accurate diagnosis and patient management.