Systematic review and meta-analysis on comparing treatment approaches of cervical fractures in ankylosing spondylitis to determine improvement in outcomes.
Amer Helal, Anne Carolus, Christopher Brenke et al.
✦ AI-curated · Sources linked
In 30 seconds
This systematic review and meta-analysis compared anterior, posterior, and combined surgical approaches for cervical fractures in patients with ankylosing spondylitis. It included 18 retrospective studies with 1,441 patients. Across nine pairwise comparisons, the only statistically significant difference was a higher complication rate with anterior than posterior surgery (OR 2.97, 95% CI 1.23-7.19, p = 0.015; I² = 65.9%). Neurological recovery and mortality did not differ significantly between approaches. The authors judged the evidence low certainty, given the retrospective designs, poor methodological quality, and substantial heterogeneity in neurological outcomes.
Key findings
- Anterior surgery was associated with more complications than posterior surgery (OR 2.97, 95% CI 1.23-7.19, p = 0.015; I² = 65.9%).
- Neurological recovery did not differ significantly in any pairwise comparison; heterogeneity was high for combined vs. anterior and combined vs. posterior comparisons (I² = 91.9% and 89.7%).
- Mortality did not differ significantly across comparisons of anterior, posterior, and combined approaches.
- All 18 included studies were retrospective, and their mean CMS was 37.89 ± 5.71, indicating poor methodological quality.
Why it matters
Cervical fractures in ankylosing spondylitis present a choice among surgical approaches, but comparative evidence remains limited. The pooled findings suggest a possible complication disadvantage for anterior versus posterior surgery, without demonstrating differences in neurological recovery or mortality. Because the evidence is low certainty, these results help frame the trade-offs but do not establish a universally superior approach.
What to keep in mind
All included studies were retrospective and had poor methodological quality. Neurological recovery comparisons showed substantial heterogeneity, limiting confidence in pooled estimates.