Study design:

Large multicenter retrospective cohort study.

Objective:

The objective of this study was to analyze the effect of fusion timing on inpatient outcomes in a nationally representative population with thoracolumbar fracture and concurrent neurological injury.

Summary of background data:

Among thoracolumbar trauma admissions, concurrent neurological injury is associated with greater long-term morbidity. There is little consensus on optimal surgical timing for these patients; previous investigations fail to differentiate thoracolumbar fracture with and without neurological injury.

Materials and methods:

We analyzed 19,136 nonelective National Inpatient Sample cases (2004-2014) containing International Classifications of Disease, Ninth Revision, Clinical Modification (ICD-9-CM) diagnosis codes for closed thoracic/lumbar fracture with neurological injury and procedure codes for primary thoracolumbar/lumbosacral fusion, excluding open/cervical fracture. Timing classification from admission to fusion was same-day, 1-2-, 3-6-, and ≥7-day delay. Primary outcomes included in-hospital mortality, complications, and infection; secondary outcomes included total and postoperative length of stay and charges. Logistic regressions and generalized linear models with gamma distribution and log-link evaluated the effect of surgical timing on primary and secondary outcomes, respectively, controlling for age, sex, fracture location, fusion approach, multiorgan system injury severity score, and medical comorbidities.

Results:

Patients undergoing surgery ≤72 hours (n=12,845) had the lowest odds of in-hospital cardiac [odds ratio (OR)=0.595; 95% confidence interval (CI), 0.357-0.991] and respiratory complications (OR=0.495; 95% CI, 0.313-0.784) and infection (OR=0.615; 95% CI, 0.390-0.969). No differences were observed between same-day (n=4724) and 1-2-day delay (n=8121) (P>0.05). Lowest odds of hemorrhage or hematoma was observed following 3-6-day delay (OR=0.467; 95% CI, 0.236-0.922). A ≥7-day delay to fusion (n=2,002) was associated with greatest odds of hemorrhage/hematoma (OR=2.019; 1.107-3.683), respiratory complications (OR=1.850; 95% CI, 1.076-3.180), and infection (OR=3.155; 95% CI, 1.891-5.263) and greatest increases in mean postoperative length of stay (4.26% or 35.3% additional days) and charges (163,562 or 71.7% additional US dollars) (P<0.001).

Conclusions:

Patients with thoracolumbar fracture and associated neurological injury who underwent surgery within 3 days of admission experienced fewer in-hospital complications. These benefits may be due to secondary injury mechanism avoidance and earlier mobilization.

Level of evidence:

Level III.