Prospective observational study of acute postlumbar laminectomy MRI

Patients who develop a lower-extremity neurological deficit after lumbar laminectomy present a diagnostic dilemma. In the setting of a neurological deficit, some surgeons use MRI to evaluate for symptomatic compression of the thecal sac. The authors conducted a prospective observational cohort study in patients undergoing open lumbar laminectomy for neurogenic claudication to document the MRI appearance of the postlaminectomy spine and to determine changes in thecal sac diameter caused by the accumulation of epidural fluid.

Eligible patients who were candidates for open lumbar laminectomy for neurogenic claudication at a single neurosurgical center between August 2007 and June 2009 were enrolled. Preoperative and postoperative MRI of the lumbar spine was performed on the same MRI scanner. Postoperative MRI studies were completed within 36 hours of surgery. Routine clinical and surgical data were collected at the preoperative visit, during surgery, and postoperatively. Images were interpreted for the signal characteristics of the epidural fluid and for thecal diameter (region of interest [ROI]) by 2 blinded neuroradiologists.

Twenty-four patients (mean age 69.7 years, range 30-83 years) were enrolled, and 20 completed the study. Single-level laminectomy was performed in 6 patients, 2-level in 12, and 3-level in 2. Preoperative canal measurements (ROI) at the most stenotic level averaged 0.26 cm(2) (range 0.0-0.46 cm(2)), and postoperative ROI at that same level averaged 0.95 cm(2) (range 0.46-2.05 cm(2)). The increase in ROI averaged 0.69 cm(2) (range 0.07-1.81 cm(2)). Seven patients (35%) had immediate postoperative weakness in at least 1 muscle group graded at 4+/5. The decline in examination was believed to be effort dependent and secondary to discomfort in the acute postoperative period. Those with weakness had smaller increases in ROI (0.51 cm(2)) than those with full strength (0.78 cm(2), p = 0.1599), but none had evidence of worsened thecal compression. On the 1st postoperative day, 19 patients were at full strength and all patients were at full strength at their 15-day follow-up. The T1-weighted epidural fluid signal was isointense in 19 of the 20 patients. The T2-weighted epidural fluid signal was hyperintense in 9, isointense in 4, and hypointense in 7 patients.

Immediately after lumbar laminectomy, the appearance of the thecal sac on MRI can vary widely. In most patients the thecal sac diameter increases after laminectomy despite the presence of epidural blood. In this observational cohort, a reduction in thecal diameter caused by epidural fluid did not correlate with motor function. Results in the small subset of patients where the canal diameter decreased due to epidural fluid compression of the thecal sac raises the question of the utility of immediate postoperative MRI.