C V Jaseena, Seshadri Reddy Varikasuvu, C P Heera, Pallavi Priya, Richa Chauhan, Sameer Dhingra
Thirty-seven studies were included. Summary sensitivity was 0.966 (95% CI: 0.92-0.98), and the specificity was 0.945 (95% CI: 0.92-0.96). The AUC was 0.979, indicating high diagnostic accuracy. Subgroup analyses revealed species-specific variations, particularly between Enterobacterales and non-fermenters. Substantial unexplained heterogeneity was present (I2 = 86.1%) for sensitivity. Among Enterobacterales, sensitivity and specificity were consistently high, with low very major error rates.
BACKGROUND: Polymyxins are considered last-resort agents for treating multidrug-resistant Gram-negative infections; however, resistance is increasing worldwide. Broth microdilution is the standard reference method for colistin susceptibility testing; however, its technical complexity and prolonged turnaround time have prompted the development of the rapid polymyxin test. The clinical performance of these rapid tests across different bacterial species remains unclear.
OBJECTIVES: To evaluate the diagnostic accuracy of the rapid polymyxin test for detecting polymyxin resistance in Gram-negative bacteria.
METHODS: Data sources: PubMed, Scopus, Web of Science, and Embase (inception to 31 August 2025).
STUDY ELIGIBILITY CRITERIA: Studies evaluating rapid polymyxin tests against broth microdilution.
PARTICIPANTS: Gram-negative bacterial isolates.
INDEX TEST: Rapid polymyxin-based assays. Reference standard: Broth microdilution. Assessment of risk of bias: QUADAS-2 tool. Methods of data synthesis: Bivariate random-effects model for summary sensitivity and specificity; hierarchical summary receiver operating characteristic curves; subgroup analyses by organism type.
PROSPERO: CRD420251140151.
RESULTS: Thirty-seven studies were included. Summary sensitivity was 0.966 (95% CI: 0.92-0.98), and the specificity was 0.945 (95% CI: 0.92-0.96). The AUC was 0.979, indicating high diagnostic accuracy. Subgroup analyses revealed species-specific variations, particularly between Enterobacterales and non-fermenters. Substantial unexplained heterogeneity was present (I2 = 86.1%) for sensitivity. Among Enterobacterales, sensitivity and specificity were consistently high, with low very major error rates.
DISCUSSION: The rapid polymyxin test demonstrates high diagnostic accuracy, particularly for Enterobacterales, but substantial heterogeneity and limited data for non-fermenters warrant caution. Implementation should be accompanied by confirmatory testing, especially in settings with a high prevalence of non-fermenting Gram-negative bacilli.