Meta-analysis of the diagnostic value of polymerase chain reaction-based nucleic acid detection methods for pertussis.

INTRODUCTION

Pertussis is an acute respiratory disease primarily caused by Bordetella pertussis, and its incidence has shown a resurgence. Polymerase chain reaction (PCR) is a recognized molecular detection method, and various related detection methods have emerged.

METHODS

This study is performed according to the PRISMA checklist. Relevant literature is adopted from the online databases. The quality of the included studies is assessed based on the REVMAN software (version 5.4). Statistical analysis is performed by STATA 14.0 and Meta-DiSc 1.4 software.

RESULTS

The results of this meta-analysis show that four different PCR techniques-conventional PCR, real-time quantitative PCR (Q-PCR), multiplex PCR, and nested PCR-have a pooled sensitivity of 0.95 (95 % CI: 0.93-0.97), pooled specificity of 0.93 (95 % CI: 0.89-0.95), pooled positive likelihood ratio (PLR) of 12.22 (95 % CI: 8.62-17.33), pooled negative likelihood ratio (NLR) of 0.05 (95 % CI: 0.03-0.08), and pooled diagnostic odds ratio (DOR) of 244.88 (95 % CI: 133.34-449.72). Multiplex PCR has the highest pooled specificity, PLR, and DOR. Nested PCR has the highest sensitivity and the lowest NLR. PCR testing holds significant value in the early diagnosis of pertussis. Multiplex PCR, nested PCR, and Q-PCR outperform conventional PCR in diagnostic performance.

CONCLUSIONS

Each type of PCR test has its advantages and disadvantages, and clinical selection should consider factors such as testing purpose, personnel, and conditions.