Background: Evaluation of ultrasound (US)-guided disc block used to diagnose discogenic pain, as described in case reports. The study aimed to ascertain the noninferiority of US-guided lumbar disc block to conventional discography in the diagnosis of discogenic low back pain (DLBP). Methods: The reports of 418 patients undergoing lumbar fusion for DLBP were stratified into a US group receiving US-guided lumbar disc block and a control group receiving fluoroscopy (FL)-assistant discography via a propensity-score matched method in a 1 : 1 ratio. The primary endpoint was the confirmatory rate defined as the rate of clinical success following surgery measured by a numerical pain rating scale score <= 2 and an Oswestry Disability Index score <= 15 at the 1-month follow-up point. Secondary outcomes included needle insertions until contrast given, procedure time, radiation dosages and adverse events. Results: The confirmatory rates for disc block and discography were 71.8% and 73.2% (difference = -1.3%, 95% confidence interval [CI]: -9.9%, 7.2%, P = 0.353). The lower bound of 95% CI did not cross the noninferiority margin of 10%. There were fewer times (8.94 +/- 2.28 vs. 16.13 +/- 3.39 min, P < 0.001) and lower radiation dosage (1689.56 +/- 898.54 vs. 8293.50 +/- 1039.09 mu Gy m(2), P < 0.001) in the US group than the control group. No serious adverse events were observed. Conclusions: US-guided lumbar disc block was not inferior to conventional discography as a diagnostic modality in the evaluation of DLBP being considered for surgery. Given that the sonographic method provided advantages in terms of facilitation of needle insertion, reduced procedure time, and attenuated radiation exposure, it might be an alternative option for surgery decision making.