Electromyographic technique for lumbar multifidus examination: Comparison of previous techniques used to localize the multifidus

Objectives: To verify and compare established techniques for needle localization in the multifidus muscle and to explore more practical techniques. Design: Human cadaver study. Setting: Anatomy laboratory in a university setting. Cadavers: Six fresh human cadavers. Intervention: A 22-gauge needle was inserted into the multifidus muscle fascicle of 6 cadavers using 2 different techniques described previously in the electrodiagnostic literature by Haig and Stein and colleagues. A mixture of colored latex and contrast dyes (0.1mL) was injected bilaterally into each fascicle at levels L1 to L5. Two electromyographers performed injections into 60 targeted muscles, affording 120 total insertions. Separate investigators dissected the muscles to determine dye position. Main Outcome Measures: Not applicable. Results: A total of 88 (73%) and 79 (66%) injections were successfully delivered to the targeted multifidus muscles using the Haig and the Stein techniques, respectively. With the Haig method, 22 injections (18%) were delivered to different superficial muscles. With the Stein method, 24 injections were delivered to a common tendon and 3 injections were delivered to the spinal canal. Conclusions: This study highlights the nonoptimizing accuracy of previous techniques for multifidus needle electromyography. A modified Haig method involving less acute needle angulation relative to the skin surface and closer insertion from the midline may increase accuracy and safety.