We conducted a meta-analysis comparing the invasiveness of automatic lancet devices, which can collect adequate amount of blood at shallow puncture depths, with conventional manual lance devices (lancet or needle) to statistically identify less invasive instruments for neonatal heel lance. We searched PubMed, Cochrane Central Register of Controlled Trials (CENTRAL), CINAHL, and Ichushi databases for studies comparing the invasiveness between automatic lancet and manual lancet or needle in term and preterm neonates in neonatal intensive care unit. This review included 9 out of 449 searched articles, with 673 neonates. Automatic lancet had significantly lower pain scores (standardised mean difference: -2.0, 95% confidence interval: -3.3 to -0.7), heart rate (mean difference: -8.0, 95% confidence interval: -13.8 to -2.1), cry duration (mean difference: -21.5, 95% confidence interval: -32.5 to -10.4), number of punctures (mean difference: -0.6, 95% confidence interval: -1.1 to -0.2), and duration of procedures (mean difference: -37.7, 95% confidence interval: -75.2 to -0.2) than manual lancet or needle. Furthermore, peripheral oxygen saturation was significantly higher in automatic lancet than in manual lancet or needle (mean difference: 4.5, 95% confidence interval: 0.5-8.5). Automatic heel lancet devices were less invasive than manual heel lance devices (lancet or needle).