Fast but accurate: a systematic review and meta-analysis on diagnostic performance of MRSA detection in clinical...

Methicillin-resistant Staphylococcus aureus (MRSA) poses a significant global health threat due to its multidrug resistance and association with severe infections. Conventional culture methods are time-consuming, usually requiring 48-72 h to obtain results, while conventional molecular methods such as PCR or qPCR, though faster, still require trained personnel and specialized instruments, which may delay timely clinical treatment and infection control. CRISPR-based methods have emerged as promising alternative tools for MRSA detection, but their real-world performance still requires comprehensive assessment. This meta-analysis aimed to systematically evaluate the diagnostic accuracy and timeliness of CRISPR/Cas systems for MRSA detection in clinical samples.

A systematic search of PubMed, Embase, Web of Science, and Cochrane Library was conducted using search terms related to MRSA, CRISPR/Cas, diagnostic accuracy, and rapid detection. Studies reporting sensitivity and specificity with extractable 2 × 2 contingency tables were included. Quality was assessed via QUADAS-2. Meta-disc 1.4.0 and Stata 16.0 were used for statistical analysis, including pooled sensitivity, specificity, likelihood ratios, diagnostic odds ratios (DOR) and summary receiver operating characteristic (SROC). Median detection time and subgroup analyses were also conducted.

Twelve studies were included. The results showed that the CRISPR-based methods showed a pooled sensitivity of 99% (95% CI: 97-100%) and specificity of 100% (95% CI: 99-100%), with a PLR of 32.68 (95% CI: 15.45-69.15), NLR of 0.03 (95% CI: 0.02-0.07), and DOR of 664.25 (95% CI: 234.59-1880.84). The median detection time across included studies was 60 min (IQR: 41.25-98.75 min).

CRISPR-based molecular assays demonstrated exceptional accuracy and rapid detection capability for MRSA in clinical settings, significantly outperforming conventional methods. However, potential publication bias and methodological limitations warrant cautious interpretation of these results.

PROSPERO ID: CRD420251115439.