BACKGROUND CONTEXTThe use of recombinant human bone morphogenetic proteins-2 (rhBMP-2) for spinal fusion has been reported to be effective. However, most studies have focused on posterolateral and anterior lumbar interbody fusion, and few have investigated posterior lumbar interbody fusion (PLIF). PURPOSEThis study aimed to determine the effectiveness and safety of the delivery of Escherichia coli-derived rhBMP-2 (E.BMP-2) with hydroxyapatite (HA) and β-tricalcium phosphate (β-TCP) poloxamer hydrogel composite carriers for PLIF. STUDY DESIGNA retrospective study. PATIENT SAMPLEPatients who underwent 1 to 3 levels of PLIF for lumbar degenerative disc disorders between 2015 and 2020 with a follow-up of ≥1 year were enrolled. In total, 254 patients (357 levels) were included in the analysis. The evaluation was performed at each segment level. In the E.BMP-2 group, 160 patients (221 levels) received autologous local bone with E.BMP-2 (maximum 0.5 mg/level), and in the control group, 94 patients (136 levels) received only local bone graft. OUTCOME MEASURESThe primary outcome of this study was to compare the X-ray and CT fusion rates between the two groups. Secondary outcomes included analysis of the patients’ clinical outcomes and postoperative complications on CT scans. METHODSClinical evaluations were performed using a visual analog scale for back pain, the Oswestry Disability Index for disability, and physical and mental component summaries of the Short Form 36-Item Form Health Survey to assess functional effects and quality of life. The fusion was evaluated using radiography and CT. On radiography, solid fusion was defined when the difference between extension and flexion was less than 5°. On CT, solid fusion was defined when the upper and lower vertebral bodies were connected by the trabecular bone (bone bridge formation). In addition, complications such as osteolysis, cage subsidence, and screw loosening were investigated using CT. RESULTSAll clinical results for low back pain, disability, and quality of life in both groups were excellent and showed statistically significant improvements compared with baseline (p<.0001). According to the X-ray evaluations, fusion was achieved in 92.31% (204/221) of the patients in the E.BMP-2 group and 82.35% (112/136) of the patients in the control group (p=.0041). According to the CT evaluations, the fusion rates were 93.21% (206/221) and 88.24% (120/136) in the E.BMP-2 and control groups (p=.1048), respectively. Except for screw loosening, which had a significantly higher incidence in the control group (p=.0014), the rates of most postoperative complications were not significantly different between the groups. CONCLUSIONSThis study demonstrated that the adjunctive use of a low dose of E.BMP-2 with HA and β-TCP hydrogel can effectively promote bone fusion, making it a promising option for patients with limited autograft availability or compromised bone quality in PLIF.