Surgical site infection (SSI), particularly deep SSI, is one of the most serious complications after spinal surgery. evaluating the risk of SSI and, correspondingly, prescription of prophylactic measures are extremely important to prevent SSI and avoid potentially devastating consequences. A retrospective study was conducted aiming to develop a point-based prediction model of deep surgical site infection in patients receiving open posterior instrumented thoracolumbar surgery.

Data of 3,419 patients in 4 hospitals from Jan 1, 2012 to Dec 30, 2021 were retrospectively collected and were evaluated aiming to develop a point-based prediction model of deep surgical site infection in patients receiving open posterior instrumented thoracolumbar surgery. Clinical knowledge-driven, data-driven and decision tree model was used to identify predictive variables of deep SSI. Internal validation was performed by using bootstrapping methods.

Inclusion Criteria:

patients who previously received open posterior instrumented thoracolumbar surgery

Exclusion Criteria:

1. were aged < 18 years;
2. underwent surgery for the spinal infection, spine revision surgery, or cervical operation;
3. were diagnosed as superficial SSI;
4. died in hospital;
5. diagnosis of SSI was not determined;
6. had missing data more than 10% were excluded.

Study population included patients who previously receceived open posterior instrumented thoracolumbar surgery due to spine disease (except that spine infection or revision surgery)