The latest meta-analysis indicated potential survival benefits from ultra-short-acting β-blockers in septic patients with persistent tachycardia. However, subsequent multicenter randomized controlled trials (RCTs) have reported conflicting findings, prompting the need for an updated meta-analysis to incorporate these newly published RCTs. Does the use of ultra-short-acting β-blockers (esmolol or landiolol) in septic patients with persistent tachycardia improve mortality? We conducted an updated systematic search till April 2nd, 2024 exploring MEDLINE, Cochrane Central Register of Controlled Trials, and Embase for RCTs reporting mortality in adult septic patients treated with esmolol or landiolol as compared to none or placebo, and published in English. Meta-analyses were conducted with the random effect model. The primary outcome was mortality at the longest follow-up, with subgroup analysis separating single from large multicenter RCTs. Eight RCTs (885 patients) were included in the primary analysis. Ultra-short acting β-blockers did not significantly improve mortality at the longest follow-up (risk ratio [RR] 0.84; 95% confidence interval [CI], 0.68-1.02; p=0.08, I2=51%, very low certainty of the evidence) and 28-day mortality (RR 0.77; 95%CI 0.59-1.00; p=0.05; I2=62%). Subgroup analyses of mortality outcomes pointed towards different results between single-center and multicenter RCTs. Trial sequence analyses (TSAs) showed that both mortality outcomes were not robust. The sensitivity analyses suggested a significant reduction in mortality by adding RCTs published in non-English language. In this updated meta-analysis, the use of esmolol or landiolol did not reduce mortality in septic patients with persistent tachycardia. However, results were not robust and outcomes differed between single-center and multicenter RCTs. Moreover, sensitivity analyses showed the fragility of the primary outcome. Further studies regarding ultra-short-acting β-blockers with advanced cardiac monitoring or serial echocardiography are warranted. The protocol was preventively registered in the PROSPERO database on the 31th January 2024 (Registration number: CRD: 42024503570).