Visual loss after corrective surgery for pediatric scoliosis: incidence and risk factors from a nationwide database.

BACKGROUND CONTEXT: Perioperative visual loss (POVL) after spinal deformity surgery is an uncommon but severe complication. Data on the incidence and risk factors of this complication after corrective surgery in the pediatric population are limited. PURPOSE: The present study aimed to investigate nationwide estimates of POVL after corrective surgery for pediatric scoliosis. STUDY DESIGN: This is a retrospective study that uses a nationwide database. PATIENT SAMPLE: The sample includes 42,339 patients under the age of 18 who underwent surgery for idiopathic scoliosis. OUTCOME MEASURES: The outcome measures were incidence of POVL and risk factors. METHODS: Patients under the age of 18 who underwent elective surgery for idiopathic scoliosis between 2002 and 2011 were identified using the Nationwide Inpatient Sample database. The incidence of POVL (ischemic optic neuropathy, central retinal artery occlusion, or cortical blindness) was estimated after application of discharge weights. Demographics, comorbidities, and operative parameters were compared between patients with and without visual loss. A multivariate logistic regression was performed to identify significant risk factors for POVL development. No funds were received in support of this work. RESULTS: The incidence of POVL was 1.6 per 1,000 procedures (0.16%). Patients with visual loss were significantly more likely to be younger and male, have Medicaid as insurance, and undergo fusion of eight or more spinal levels compared with patients without visual loss. Following multivariate analysis, older patients (odds ratio [OR]: 0.84; 95% confidence interval [CI]: 0.77-0.91) and female patients (OR: 0.08; 95% CI: 0.04-0.14) were significantly less likely to develop POVL compared with younger and male patients. On the other hand, having Medicaid as insurance (OR: 2.13;95% CI: 1.32-3.45), history of deficiency anemia (OR: 8.64; 95% CI: 5.46-14.31), and fusion of eight or more spinal levels (OR: 2.40; 95% CI: 1.34-4.30) were all independently associated with POVL. CONCLUSIONS: In this nationwide study, the incidence of POVL after scoliosis surgery in patients under the age of 18 was estimated at 0.16%, similar to the rate reported in adult patients. Cortical blindness accounted for all cases of POVL in the present study. Younger patients, patients with history of deficiency anemia, and patients undergoing long-segment fusions may be at increased risk of POVL after corrective surgery for pediatric scoliosis.

Duke Scholars

Author Christopher Ignatius Shaffrey Orthopaedic Surgery