Effects of preoperative embolization on outcomes in histopathologically nonhypervascular spinal metastases: a propensity score-matched study.

OBJECTIVE

While preoperative embolization (PrE) is known to reduce blood loss during spinal surgery for histopathologically hypervascular metastases, its efficacy in nonhypervascular spinal metastases remains underexplored. This study aimed to evaluate the effectiveness of PrE in patients with nonhypervascular spinal metastatic tumors, focusing primarily on estimated blood loss (EBL) and secondarily on perioperative outcomes.

METHODS

This retrospective study included 152 patients diagnosed with nonhypervascular thoracolumbar spinal metastases who underwent surgery between January 2018 and December 2022. Propensity score matching was performed to balance surgical indications and clinicodemographic characteristics, resulting in 55 matched pairs (110 patients) with or without PrE. Surgical outcomes (overall EBL, perioperative blood transfusion, operation time, reoperation rate, massive EBL [defined as ≥ 2500 mL], and 30- and 90-day mortality) were compared. Prespecified subgroup analyses were also conducted.

RESULTS

The matched PrE group had significantly lower overall EBL (median 600 [IQR 300-1200] mL) compared with the matched non-PrE group (median 900 [IQR 500-1800] mL) (p = 0.02). The incidence of massive EBL was also lower in the PrE group (3 patients [5.5%]) than in the non-PrE group (10 patients [18.2%], p = 0.03; OR 0.26 [95% CI 0.06-0.91]). No significant differences were observed between groups regarding perioperative transfusion, operation time, reoperation rate, 30-day mortality, or 90-day mortality. However, 1 case of PrE-related spinal cord infarction occurred. Subgroup analyses revealed that PrE was more effective in reducing massive EBL among patients with hyperenhancement on preoperative CT-digital subtraction angiography and those undergoing highly invasive surgery (pinteraction = 0.02 and 0.03, respectively).

CONCLUSIONS

After adjusting for clinicodemographic factors, PrE was associated with reduced EBL in patients with nonhypervascular thoracolumbar spinal metastases. Moreover, patients with radiological hyperenhancement or undergoing highly invasive surgery may derive greater benefit from PrE in mitigating massive EBL.