Usefulness of clinical predictors for preoperative screening of deep vein thrombosis in hip fractures

Background: Recent studies showed that preoperative deep vein thrombosis (DVT) was common after hip fracture (HF), and preoperative DVT screening has been recommended for preventing the fatal DVT-related complications, especially in elderly HF patients with high surgical risk. However, to our knowledge, no previous studies have demonstrated the correlation between the clinical risk predictors and preoperative DVT. Therefore, this study aimed to correlate those clinical predictors related to DVT risk assessment with the incidence of preoperative DVT. Methods: A prospective study was conducted, between July 2015 and June 2016, in 92 HF patients. All patients were evaluated for the DVT-related risk, as patients' characteristics, clinical signs, D-dimer, DVT risk assessment score (Wells score and Caprini score), and then underwent doppler ultrasonography preoperatively. The incidence of preoperative DVT was correlated with each clinical risk predictor, and then significant factors were calculated for diagnostic accuracy. Results: The average patients' age was 78 +/- 10 years. Sixty-eight patients (74%) were female. The incidence of preoperative DVT was 16.3% (n = 15). The median time from injury to doppler ultrasonography was 2 days (range 0-150 days). DVT group showed a significantly higher in Wells score and Caprini score compared to the non-DVT group (p < 0.05 all). Sensitivity and specificity of Wells score >= 2 and Caprini score >= 12 were 47 and 81, and 93 and 35%, respectively. Conclusion: DVT risk assessment may be helpful for stratifying the risk of preoperative DVT in elderly HFs. Those with Caprini score >= 12 should be screened with doppler ultrasonography preoperatively. Those with Wells score 0-1 had low risk for preoperative DVT, so the surgery could perform without delay.