<p>While botulinum neurotoxin type A (BoNT-A) has demonstrated efficacy in the treatment of blepharospasm and hemifacial spasm, comparative efficacy and safety of different BoNT-A formulations remain uncertain. We conducted a systematic review and network meta-analysis with random-effects Bayesian models of randomized controlled trials comparing BoNT-A preparations with each other or with placebo. Separate networks were constructed for duration of clinical effect, analyzed as mean differences (MD), and ptosis, analyzed as odds ratios (ORs). The network included 13 randomized controlled trials with 870 participants contributing to duration outcomes and 964 participants to ptosis outcomes. All BoNT-A preparations demonstrated a significantly longer duration of clinical effect compared with placebo in a combined blepharospasm and hemifacial spasm population. No clinically meaningful differences in duration were observed between active preparations. AbobotulinumtoxinA (Abo/BoNT-A), onabotulinumtoxinA (Ona/BoNT-A), incobotulinumtoxinA (Inco/BoNT-A) and lanbotulinumtoxinA (Lan/BoNT-A) were associated with significantly higher odds of ptosis compared with placebo. Abo/BoNT-A, which was generally administered at higher dose ratios in the included studies, showed higher odds of ptosis than Ona/BoNT-A, although estimates were imprecise with wide credible intervals. In the dose-level network, higher-dose BoNT-A was associated with a greater risk of ptosis than lower-dose BoNT-A. The available evidence suggests that BoNT-A formulations have comparable efficacy in the combined blepharospasm and hemifacial spasm population. Differences in ptosis risk may reflect differences in dosing strategies rather than intrinsic differences between BoNT-A formulations. The review protocol was registered in PROSPERO (CRD420261299121).</p>