The Incidence of Intravascular Penetration in Medial Branch Blocks: Cervical, Thoracic, and Lumbar Spines

Spine (Phila Pa 1976). 2008 Mar 15;33(6):E174-7. doi: 10.1097/BRS.0b013e318166f03d.


Study design: Clinical observational study.

Objective: To quantify the incidence of inadvertent intravascular injections in spinal medial branch blocks in a clinical setting.

Summary of background data: Previous research established the rate of inadvertent intravascular injection in lumbar medial branch blocks at 8%. The incidence of intravascular injection in cervical and thoracic medial branch blocks has not been reported previously. This study establishes the rate of inadvertent intravascular injection in patients receiving medial branch blocks of the cervical and thoracic spines. Further, this study reports a significantly lower rate of inadvertent intravascular injection for lumbar medial branch blocks than previously reported.

Methods: Patients were originally referred to the clinic, for diagnosis and treatment of chronic spinal origin somatic pain. Medial branch blocks were then performed as diagnostic procedures to confirm the zygapophysial joint(s) as the suspected source of pain. Blocks were performed by experienced practitioners on nonidentified patients over a 3-year period. Clinical observations were recorded for 14,312 separate medial branch block levels. The level of the spine and the incidence of inadvertent intravascular injections were recorded.

Results: This study demonstrates that the overall incidence of intravascular penetration in medial branch blocks is rare, with an overall rate of 3.5%. This study also establishes the rate of intravascular injection for levels within the spine: the cervical spine is likely to be intravascular 3.9% of the time and the lumbar spine 3.7%, whereas the thoracic spine is significantly lower, with just 0.7% injections reported as intravascular. Significant differences were also observed between individual vertebral levels.

Conclusion: The false-negative rate for medial branch blocks is likely to be lower than previously reported. The rate of inadvertent intravascular injection for thoracic medial branch blocks is 0.7%. Cervical and lumbar medial branch blocks are associated with an overall rate of 3.9% and 3.7%, respectively. Although these rates are lower than previously reported, the incidence of false-negative blocks still justifies the use of contrast to confirm nonvascular injection.

Publication types

  • Comparative Study

MeSH terms

  • Anesthesia, Spinal / adverse effects*
  • Cervical Vertebrae / blood supply*
  • Cervical Vertebrae / drug effects
  • Humans
  • Incidence
  • Injections, Intra-Arterial
  • Injections, Intravenous
  • Lumbar Vertebrae / blood supply*
  • Lumbar Vertebrae / drug effects
  • Medical Errors* / trends
  • Nerve Block / adverse effects
  • Thoracic Vertebrae / blood supply*
  • Thoracic Vertebrae / drug effects