# Comparison of Phenotypic and Genotypic Detection of Drug Resistance in Acinetobacter baumannii in a Tertiary Care Hospital

## TL;DR

This study compares how well genetic and traditional lab tests detect drug resistance in a dangerous hospital bacteria.

## Contribution

The paper compares molecular and phenotypic methods for detecting drug resistance in Acinetobacter baumannii.

## Key findings

- 93 out of 104 A. baumannii isolates were drug-resistant.
- Phenotypic methods detected MBL production in 36.54%-89.42% of isolates.
- Molecular detection identified resistance genes in 60% of isolates.

## Abstract

Background

Acinetobacter baumannii (A. baumannii) is a common cause of nosocomial infection. Multidrug-resistant A. baumannii is a life-threatening and therapeutic challenge, especially in critically ill and vulnerable patients. Drug resistance in A. baumannii is conferred by various underlying mechanisms. This prospective cross-sectional study aims to study the comparison of the phenotypic MBL-E test and molecular tests conferring drug resistance to A. baumannii.

Materials and methods

Different clinical samples were collected in a time period of two years. Isolated A. baumannii strains were studied for the drug-resistance profile by the Kirby disc method. These drug-resistant isolates were further subjected to metallo-beta-lactamase (MBL) production by molecular detection of OXA-48, NDM, and VIM genes and phenotypic methods by the double-disk synergy test, modified Hodge test, and MBL-E test.

Results

A total of 104 A. baumannii isolates were obtained from 3965 samples. Ninety-three (89.4%) of these 104 isolates were found to be drug-resistant which were further analyzed by phenotypic methods for MBL production which showed a detection range of 36.54%-89.42% as compared to a molecular method where detection was observed as 56 (60%).

Conclusion

Molecular detection of drug-resistance conferring genes can be a time-effective method as compared to phenotypic detection. However, genetic methods have their own limitation and additional research empaneling in a single test is required.