AUTHOR=Tongta Saowalak , Angsnuntsukh Chanika , Saisongcroh Tanyawat , Woratanarat Thira , Tangsopa Yaowaret , Woratanarat Patarawan 

TITLE=Deep vein thrombosis screening in pediatric orthopedic patients

JOURNAL=Frontiers in Surgery

VOLUME=Volume 10 - 2023

YEAR=2023

URL=https://www.frontiersin.org/journals/surgery/articles/10.3389/fsurg.2023.1041578

DOI=10.3389/fsurg.2023.1041578

ISSN=2296-875X

ABSTRACT=Background: Deep vein thrombosis (DVT) is an important clinical condition that leads to subsequent morbidity and mortality in children, particularly those who involve with operative procedures. The preoperative assessment for DVT in children may vary among different population risk factors, and types of surgery. This study was aimed to evaluate screening methods for DVT in pediatric orthopaedic service.
Method: We performed a retrospective cohort study of orthopaedic patients aged < 18 years at Ramathibodi Hospital, Bangkok, Thailand from 2015-2019. The inclusion criteria were children scheduled for orthopaedic surgery; performed D-dimer test, Wells score, and Caprini score; and underwent Doppler ultrasonography for DVT screening. The exclusion criteria were incomplete data or inconclusive ultrasonographic results. Age and results of D-dimer test, Wells score, and Caprini score were collected in all patients. The outcome assessment was ultrasound proven DVT. Screening abilities of each test were analyzed as sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), likelihood ratio (LR) for positive and negative tests, and area under the receiver operating characteristic curve (AUC).
Results: There were 419 children included in the study. Five (1.19%) patients were diagnosed with DVT. The mean age was 10.16 ± 4.83 years. The D-dimer ≥ 500 ng/ml had sensitivity 100% (95%CI 47.8%–100%), specificity 36.7% (95%CI 32.1%–41.6%), PPV 1.9% (95% CI 0.6-4.3%), and NPV 100% (95% CI 97.6-100%). The Wells score ≥ 3 demonstrated the sensitivity 0% (95% CI 0–52.2%), specificity 99.3% (95% CI 97.9-99.9%), and LR for a negative test 1.00 (95% CI 1.00-1.01). The Caprini score ≥ 11 had the sensitivity 0% (95% CI 0-52.2%) and specificity 99.8% (95% CI 98.7–100%). The parallel test included either D-dimers ≥ 500 ng/ml or Wells score ≥3 or Caprini score ≥11 points generated the sensitivity 100% (95% CI 47.8-100%), specificity 36.7% (95% CI 32.1-41.6%), LR for a positive test 1.58 (95% CI 1.47-1.70), and AUC 0.68 (95% CI 0.66–0.71).
Conclusions: D-dimer test exhibited moderate ability in predicting the development of DVT among pediatric orthopaedic surgery. The Wells score and Caprini score had low performance in identifying the hospitalized children at increased risk of DVT events.