Nomogram for Predicting Spinal Surgery Site Infection Risk, 2017-2024

A multicenter retrospective study of 1,049 patients who underwent spinal surgeries between 2017 and 2024, with 59 cases of surgical site infection. Author Yang Sun developed a nomogram prediction model based on six independent risk factors identified through regression analysis.

Use Cases

• Training a binary classification model to predict surgical site infection risk based on patient age, diabetes status, and albumin levels.
• Validating clinical prediction tools for postoperative complications using a dataset with known infection outcomes.
• Analyzing the relationship between operative factors like blood loss and operative time on infection risk following spinal surgery.

Strengths

• Dataset includes 1,049 patient records from a multicenter study.
• Model was internally validated with a concordance index of 0.787, indicating good discrimination.
• Six statistically significant independent predictors are explicitly identified with odds ratios.

Limitations

• Row count for the underlying data table is unknown, limiting suitability assessment.
• Column-level documentation is absent; field semantics must be inferred after download.
• The dataset is very small at 17.7 KB, indicating limited scope.

Provenance

Source

figshare, author Yang Sun.

Collection Method

Retrospective data collection from patients who underwent spinal surgeries.

Time Range

2017 to 2024

Freshness

Last updated 2026-04-28 04:24:43; freshness should be verified.

Geography

null