4:10 Instrumented posterior occipitocervical fusion in inflammatory arthritis

Abstract

Purpose of study: This study evaluates the outcome of posterior instrumented occipitocervical fusion (PIOF) in an inflammatory arthritis population.Methods used: Eighteen consecutive patients undergoing PIOF were evaluated. Patient demographics and joint counts, Ranawat grades (RGs) and neck pain visual analogue scores (VAS) were collected prospectively during clinic visits before and after surgery. All patients had PIOF with autograft, and six patients had a preceding transoral odontoidectomy during the same anesthetic. All patients were braced postoperatively for a 3-month duration. All patients were seen and assessed at 6 weeks, 3 months, 6 months and 1 year annually.of findings: Seventeen of the 18 patients had rheumatoid arthritis, including 1 with juvenile onset rheumatoid arthritis and 1 with psoriatic arthritis. There were 4 males and 14 females; 8 were Caucasian, 8 were Aboriginal. Mean age was 58 years (range, 32 to 85), mean disease duration was 20 years (range, 10 to 35) and mean number of joints affected was 41 (range, 16 to 71). Sixty-seven percent were taking corticosteroids. RG before surgery was 2 in 28%, 3a in 55% and 3b in 17%. Mean VAS score before surgery was 8.8 of 10. All patients had an atlantodens interval of greater than 5 mm and vertical migration of the odontoid (VMO), and six patients had posterior atlantodens interval of less than 13 mm. Six patients had a transoral odontoidectomy for an irreducible C1–C2 subluxation and VMO. All patients had a PIOF. Eight patients (44%) had fusion to C3, six (33%) to C4 and two each (11% each) to C6 and C7, respectively. Mean odds ratio (OR) time was 239 minutes (range, 100 to 600), mean blood loss was 392 ml (range, 100 to 1,000). Mean follow-up was 24 months (range, 3 to 56; there was one death at 3 months). Fourteen patients (78%) had a postoperative RG of zero (no paresthesia or weakness), and 83% had an improvement of at least two grades on the RG. The postoperative VAS declined by a mean of 7.8 (range, 3 to 10). There were three deaths at 3, 18 and 21 months, respectively. Causes of death were pneumonia with respiratory failure and two cases of cardiopulmonary failure. Two patients (11%) had a major complication (hardware failure in one, and a soft palate dehiscence and hydrocephalus in the other), and four patients (22%) had minor complications. Those patients who died had significantly worse underlying arthritis (disease duration 28 vs. 18 years: p=.017; mean affected joint count 58 vs. 38; p=.026), had worse preoperative RG (100% at least grade 3A vs. 67% at least grade 3A; p=.035) required more extensive surgery (mean OR time 447 vs. 197 minutes; p=.002; transoral approach in all three deaths vs. 3 of 15 survivors; p=.025; level of cervical fusion to C6–C7 in 67% vs. 13%; p=.043); more blood loss (617 cc vs. 347 cc; p=.047) and had less improvement in their postoperative RG (67% less than 1 grade improved vs. 7% less than 1 grade improved; p=.038). Age and corticosteroid use were not significantly associated with death.Relationship between findings and existing knowledge: This study supports previous findings that PIOF is a safe and effective procedure for inflammatory arthritis patients with craniocervical involvement, but that the long-term outcome is less favorable in patients with high preoperative RG.Overall significance of findings: Overall inflammatory arthritis patients undergoing PIOF had markedly improved RG and VAS. Although there were no perioperative deaths, patients requiring transoral odontoidectomy had worse underlying inflammatory arthritis, worse radiographic and neurologic parameters and a higher long-term risk of death.Disclosures: Device or drug: posterior cervical spine instrumentation: Synthes Starlock. Status: approved. Device or drug: posterior cervical spine instrumentation: Sofamor Danek Lateral Mass Plates. Status: approved.Conflict of interest: No conflicts.