Paired and matched analysis of neurological outcomes in revision surgery for cervical myelopathy following delayed neurological decline.

Background: Delayed neurological decline may be experienced following successful decompression surgery for cervical myelopathy. Our objective was to analyze neurological recovery upon revision surgery with relation to the index procedure and a matched control. Methods: Fourteen patients underwent both primary and revision decompression at a single academic center. Peri-operative clinical, radiological, and surgical details were retrieved. Neurological outcomes [change in modified Japanese Orthopedic Association (mJOA), recovery ratio] following the second surgery were compared to (I) the primary operation and (II) a control subject receiving primary decompression matched for gender, age, mJOA score, and surgical approach. The minimum clinically important difference (MCID) in mJOA score was set at 2.5. Results: Revision decompressions were performed 6.8±4.2 years following the index surgery, when patients were 61.4±11.0 years of age. An increase in mJOA score of 2.7±2.0 following revision surgery was similar to that achieved after the primary operation (2.2±2.1, P=0.616). A recovery ratio of 38.1%±25.4% upon revision compared favorably to that following the primary operation (35.0%±37.4%, P=0.867). Non-inferiority testing between revision surgery and the first operation (P=0.02) demonstrated a similar capacity to achieve the MCID as did comparison with matched subjects (P<0.01). Conclusions: Patients were able to make up for lost neurological gains following revision surgery. Careful selection of cases for revision likely facilitated recovery. Recovery trajectories should be consolidated upon larger sample sizes allowing for identification of prognostic factors.