Assessing Risk for Complications in Acute Hematogenous Osteomyelitis in Children: Validation of 2 Predictive Scores
Adriana Sarmiento Clemente1, J Chase McNeil1, Kristina G Hultén1
1Department of Pediatrics, Division of Infectious Diseases, Baylor College of Medicine and Texas Children's Hospital, Houston, Texas, USA.
Insights
Predictive scores for acute hematogenous osteomyelitis (AHO) complications in children show promise. The A-SCORE demonstrated strong predictive ability for acute issues, while the C-SCORE requires further validation for chronic complications.
Area of Science:
- Pediatric Orthopedics
- Infectious Diseases
- Clinical Prediction Models
Background:
- Acute hematogenous osteomyelitis (AHO) in children can lead to severe, unpredictable complications.
- Existing predictive scores (A-SCORE, C-SCORE) require external validation.
Purpose of the Study:
- To externally validate the predictive accuracy of the A-SCORE and C-SCORE for acute and chronic complications in pediatric AHO.
- To identify novel predictors for AHO complications.
Main Methods:
- Retrospective study of 418 children diagnosed with AHO.
- Analysis of the performance of A-SCORE and C-SCORE using established criteria.
- Evaluation of variables such as bone abscess, fever, arthritis, disseminated disease, CRP levels, and bacteremia.
Main Results:
- An acute complicated course occurred in 25.4% of patients.
- The A-SCORE showed high sensitivity (81%) and AUC (0.87) for predicting acute complications.
- The C-SCORE demonstrated lower specificity (86%) and AUC (0.71) for chronic complications than initially reported.
Conclusions:
- The A-SCORE and C-SCORE can aid in managing pediatric AHO and its long-term consequences.
- Prospective studies are necessary to confirm the clinical utility of these predictive scores.
Background:
Acute hematogenous osteomyelitis (AHO) can be associated with severe complications which can be difficult to predict in the clinical setting. The previously published predictive acute complication score ("A-SCORE") and chronic complication score ("C-SCORE") show promise, however, further external validation is needed.
Methods:
We performed a retrospective study of 418 children with AHO and analyzed the performance of A-SCORE (variables included bone abscess, fever after 48 h of starting antibiotics, suppurative arthritis, disseminated disease, and delayed source control) to predict risk for acute complicated course (treatment failure, prolonged admission, and/or need for ≥2 bone debridements) and C-SCORE (includes disseminated disease, bone debridement, and CRP ≥10 mg/dL at 2-4 days after starting antibiotics) to predict chronic complications (growth restriction, pathologic fracture, chronic osteomyelitis, avascular necrosis, joint deformity, and/or frozen joint).
Results:
An acute complicated course occurred in 106/418 (25.4%); 51/380 (13.5%) with complete follow-up data had a chronic complication. The A-SCORE performed with similar specificity (78%) and negative predictive value (NPV) (92%), and higher sensitivity (81%) and increased area under the receiver operating curve (AUC) (0.87) in our population. The C-SCORE performed with similar sensitivity (64%) and NPV (94%) but had lower specificity (86%) and AUC (0.71) than originally reported. Other variables associated with development of complications such as tibia involvement and bacteremia ≥2 days were identified but did not result in significantly improved predictive scores.
Conclusions:
Predictive A-SCORE and C-SCORE for AHO complications in children may help guide acute management and long-term follow-up decisions. Prospective studies are needed to determine their applicability.
