Diabetes predisposes to spine degenerative diseases often requiring surgical intervention. However, the statistics on the prevalence of spinal fusion success and clinical indications leading to the revision surgery in diabetes are conflicting. The purpose of the presented retrospective observational study was to determine the link between diabetes and lumbar spinal fusion complications using a database of patients (n = 552, 45% male, 54 ± 13.7 years old) residing in the same community and receiving care at the same health care facility. Outcome measures included clinical indications and calculated relative risk (RR) for revision surgery in diabetes. A paravertebral tissue recovered from non-union site of diabetic and non-diabetic patients was analyzed for microstructure of newly formed bone. Diabetes increased RR for revision surgery due to non-union complications (2.80, 95% CI [1.12, 7.02]) and degenerative processes in adjacent spine segments (2.26, 95% CI [1.45, 3.53]). In diabetes, a risk for revision surgery exceeded the RR for primary spinal fusion surgery by 44% (2.36, 95% CI [1.58, 3.52] vs 1.64, 95% CI [1.16, 2.31]), which was already 2-fold higher than diabetes prevalence in the studied community. Micro computed tomography of bony fragments found in the paravertebral tissue harvested during revision surgery revealed structural differences suggesting that newly formed bone in diabetic patients may be of compromised quality, as compared to non-diabetic patients. In conclusion, diabetes significantly increases risk for unsuccessful lumbar spine fusion outcome requiring revision surgery. Diabetes predisposes to the degeneration of adjacent spine segments and pseudoarthrosis at the fusion sites, and affects structure of newly formed bone needed to stabilize fusion.