Lumbar spinal stenosis (LSS) causes low back pain, leg pain, numbness in the leg, and neurogenic intermittent claudication. Epidural steroid injection (ESI) has been used for treating spinal stenosis symptoms. We hypothesized that dural pulsation was variable for lumbar spinal stenosis. In cases of the presence of dural pulsation, the pain relief after the ESI was better than in the absence of dural pulsation. This study aimed at investigating the relationships between the presence or absence of spinal dural pulsations and the efficacy of ESI. A total of 71 patients were enrolled in this prospective study. Prior to the ESI, the dural pulsation was measured using a 5-1 MHz array ultrasound transducer. The visual analogue scale (VAS) score was measured pre-ESI and 2 weeks post-ESI and 4 weeks post-ESI. At 4 weeks post-ESI, dural pulsation was rechecked. The VAS scores improved after the ESI procedure regardless of the presence or absence of dural pulsation. There was a correlation between the pulsation of the dura and post-ESI VAS scores. However, VAS was not significantly different for different grades of stenosis. The ESI was effective in patients with spinal stenosis in short-term follow-up. Dural pulsation of the spinal cord was a positive predictive factor for the ESI effect, but the grade of spinal stenosis severity had no effect on the effectiveness of ESI.