Comprehensive Wound Risk Stratification of Rib-Based Distraction Instrumentation Procedures
Patrick J Cahill1, Richard E Campbell2, Nikita Lakomkin3
1Division of Orthopaedics, Children's Hospital of Philadelphia, 3401 Civic Center Blvd, Philadelphia, PA 19104, USA.
Study Design:
Single-center retrospective analysis of a prospectively collected registry.
Objectives:
Identify factors predictive of rib-based distraction (RBD) instrumentation wound complication. Create a risk stratification model for RBD instrumentation wound complication.
Summary Of Background Data:
RBD instrumentation procedures have a high rate of wound complications, often requiring unplanned operative treatment. Currently, there is a relative lack of understanding of RBD complication risk factors compared with the comprehensive understanding of complication risk factors for other spine surgeries.
Methods:
Between January 2011 and September 2015, patients treated with RBD instrumentation at a single institution were analyzed for risk factors associated with surgical wound complications that resulted in unplanned operative treatment. Univariate logistic regression determined predictors of wound complication and multivariate regression determined independent predictive factors; α = 0.10.
Results:
A total of 122 patients aged 0-18 years underwent 140 implant surgeries in which 22 resulted in complications: 18 (82%) infectious and 4 (18%) noninfectious. Mean age at surgery was 5.2 years. Univariate analysis showed a correlation between wound complication rates and the following: male gender (p = .097), diapered patient with lower back incision (p = .004), bilateral procedure (p = .008), more than three incisions (p = .011), left iliac incision (p = .097), right iliac incision (p = .009), patient age ≤4 years (p = .10), and operative time >150 minutes (p = .079). Multivariate analysis identified the following independent predictors: age ≤ 4 years (p = .002), male gender (p = .04), number of skin incisions (p = .001), left iliac incision (p = .018), and nutritionally challenged (p = .044). The multivariate model predicted wound complications with an area under the receiver operating characteristic curve of 0.88.
Conclusions:
Knowledge of risk factors for RBD instrumentation wound complications can be used to construct patient risk models. This can identify patients at higher risk for complications and influence clinical decision making.
Level Of Evidence:
Level II.
