Plain acupuncture uses manipulation (rotation or varying the depth of insertion of the needle) to increase its effect. However, in commonly used electroacupunture (EA), variable manipulations have not been used. This study was performed to investigate the possibility of an increase in analgesic effect by adding manipulation to EA. The pain index used was the Tail-Flick latency (TFL) of the rat, which was lightly anesthetized with thiopental sodium (intraperitoneally). Four types of manipulation were used. Rotation and varying the depth of the needle (RN and VN) was employed using two different types of manipulation during each 20 min stimulation of EA. Each manipulation persisted for 1 min out of every 5 min (long - duration and long - interval: LDLI) or 12 s every 1 min (short - duration and short interval: SDSI). EA produced an increase in TFL; peak value was 49.7+/-12.2% of the pre - EA and occurred immediately after cessation of 20 min of EA stimulation. Performing RN or VN combined with EA also increased TFL more than just EA and a greater peak increase in TFL was observed with a SDSI - RN and SDSI - VN as compared to a LDLI - RN and LDLI - VN (77.5+/-13.8, 79.2+/-19.8 and 67.3+/-14.0%, 65.6+/-23.7% of the pre - EA, respectively). These results indicate that manipulation combined with EA produces a more potent antinociception than when only EA is applied.