Risk factors and outcomes for catastrophic failures at the top of long pedicle screw constructs: a matched cohort analysis performed at a single center.

STUDY

DESIGN

Retrospective review with matched-cohort analysis performed at a single institution.

OBJECTIVE

To determine risk factors and outcomes for acute fractures at the proximal aspect of long pedicle screw constructs.

SUMMARY OF BACKGROUND DATA: Acute fractures at the top of long segmental pedicle screw constructs (FPSC) can be catastrophic. Substantial surgical increase in lordosis may precipitate this problem. In relation to a matched cohort, we postulated that age, body mass index (BMI), and significant correction of lumbar lordosis would increase risk of FPSC and patients with FPSC would have lesser improvements in outcomes.

METHODS

Thirteen patients who sustained FPSC between 2000 and 2007 were evaluated. During this time, 264 patients aged 40 or older had a spinal fusion from the thoracic spine to the sacrum using an all-pedicle screw construct. A cohort of 31 of these patients without FPSC but with all pedicle screw constructs was matched for diagnosis of positive sagittal imbalance, gender, preoperative C7 sagittal plumb, and number of levels fused.

RESULTS

There was a significant difference in age (P = 0.02) and BMI (P = 0.006) between the matched groups. There was no significant difference in preoperative/postoperative C7 plumb or change in lumbar lordosis between groups. Acute neurological deficit developed in 2 patients; both patients improved substantially after revision surgery. Nine patients underwent proximal extension of the fusion. For 7 of the 13 FPSC patients with bone mineral density data (BMD) available, average T score was-1.73; -0.58 for the matched group (10/31 with bone mineral density data) (P = 0.02).

CONCLUSION

Factors that increased the risk of FPSC included obesity and older age. Osteopenia increased the risk as evidenced by BMD (based on 17 patients) and the older age of these patients. There was no statistical difference in clinical improvement between groups based on ODI, but the FPSC group did demonstrate a smaller improvement in ODI score than the matched cohort.