The likelihood of treatment failure in COVID-19 patients with bacterial superinfection stems from phenotypic, viz., biofilms, and genotypic mechanisms. This cross-sectional study aimed to determine the inhibitory concentrations of quinolones—nalidixic acid, norfloxacin, ciprofloxacin, ofloxacin, and levofloxacin—in biofilm formers [minimum biofilm inhibitory concentration (MBIC)] and non-formers [minimum inhibitory concentration (MIC)] as well as correlate the quinolones’ folds with the presence of plasmid-mediated quinolone-resistance (PMQR) genes in quinolone-resistant bacteria isolated from COVID-19 inpatients. Quinolone-resistant isolates (n=193), verified through disc diffusion, were tested for quinolone inhibitory concentrations and biofilm formation with broth microdilution and microtiter plate methods, respectively. Polymerase chain reaction was used to detect PMQR genes. MIC to MBIC median increase in folds for ciprofloxacin, ofloxacin, and levofloxacin was 128 (2-8,192), 64 (4-1,024), and 32 (4-512) in gram-positive cocci (GPC), respectively, while it was 32 (4-8,192), 32 (4-2,048), and 16 (2-1,024) in fermentative-gram-negative bacilli (F-GNB), and 16 (4-4,096), 64 (2-64), and 16 (8-512) in non-fermentative-gram-negative bacilli (NF-GNB). Biofilm-forming F-GNB (32/126) and NF-GNB (10/24) harbored qnrB [11/32 versus (vs.) 3/10], aac(6')-Ib-cr (10/32 vs. 4/10), and qnrS (9/32 vs. 0/10) genes, respectively. A 32-fold median increase in ciprofloxacin was significantly associated with qnrA and qnrS in F-GNB and NF-GNB, respectively. F-GNB and NF-GNB biofilms were significantly associated with aac(6')-Ib-cr and qnrS genes, respectively. Nearly one-third of the superinfecting bacteria in COVID-19 patients formed biofilms, and had at least one PMQR gene, increasing the need for quinolone inhibitory concentrations.

Data on patients' demographics and laboratory findings, including microbiological analyses, were obtained using patient information sheets. Any missing or unclear records were clarified by communicating with involved healthcare professionals, patients, or families using the phone numbers listed in the patient information sheet. Subsequently, the study variables were recorded in Microsoft Excel, version 10.0.

The patient information sheet data was entered into Microsoft Excel version 10.0. The extracted data was imported into Statistical Package for Social Science (SPSS), version 17.0 (SPSS Inc., Chicago, Illinois, USA), for conducting statistical analysis.

Using descriptive statistics, the data were initially calculated as frequency (n) and percent (%). The quantitative variables were calculated as medians [interquartile range (IQR)]. Statistical associations between the two groups were analyzed using independent student t-tests and Chi-square tests for quantitative and qualitative variables, respectively. A p-value of <0.05 indicated a significant correlation between the variables at a 95% confidence interval.

Statistical significance had to be determined for each bacterial category [fermentative-gram-negative bacteria (F-GNB), non-fermentative-gram-negative bacteria (NF-GNB), and gram-positive cocci (GPC)], considering the distinct testing protocols of quinolones utilized for AST, MIC, or MBIC. 

Microsoft Excel version 10.0 is required to open the data file of the manuscript.