Aims To compare the anaesthetic efficacy of articaine, lidocaine and mepivacaine for buccal infiltration (BI) following lidocaine inferior alveolar nerve block (IANB) in mandibular posterior teeth with irreversible pulpitis.Materials and methods Patients with irreversible pulpitis in mandibular posterior teeth and unsuccessful IANB were randomly assigned to three groups: articaine group (n = 52), lidocaine group (n = 52) and mepivacaine group (n = 52). They were instructed to rate the pain experienced at four phases (before the injection, after IANB, after BI and during endodontic access) on a Heft-Parker visual analogue scale (VAS). Success was defined as the ability to access and instrument the tooth with no pain or mild pain (VAS rating ≤54 mm) after BI.Results Multivariate logistic regression analysis showed that articaine was associated with a higher success rate compared with lidocaine (OR = 3.89, 95% CI: 1.35-11.27; P = 0.02) and mepivacaine (OR = 3.67, 95% CI: 1.24-9.75; P = 0.01), after controlling for age, gender and initial pain. VAS ratings were significantly lower in the articaine group compared with those in the lidocaine group and mepivacaine group after BI and during endodontic access (P <0.01).Conclusion Articaine as a supplemental BI following IANB is a more successful anaesthetic agent in mandibular posterior teeth with irreversible pulpitis compared with lidocaine and mepivacaine.