| Objective:Analyze PICC-DVT risk factors of 564 cases indwelling PICC elderly inpatients according to retrospective study of them. Establish a risk prediction model based on multi-factor analysis model according to the Clinical Evaluation Method (Clinical Prediction Rules, CPRs) constructing principle, in order to provide assessment tools with higher sensitivity and specificity for clinical risk evaluation, and to enhance the level of prevention of PICC-DVT of elderly inpatients.Methods:The data of elderly inpatients who were given PICC catheters was retrieved from January 2008 to December 2015 in our hospital. According to literature and clinical experience,the data collected may be associated with PICC-DVT,such as general information,signs,examinations,laboratoryindex,complications,medication, catheteriza tion, catheter thrombosis cases etc. Then use the statistical software SPSS20.0 to analyze; statistical methods including univariateanalysis by ?2test, t-test. Multivariate logistic regression and Cox proportional hazard models were applied to analyze the effect of risk factors associated with catheter related UEDVT during follow-up.Then estimate the value of OR, HR and 95%CI.The results were corrected by clinical factors for assignment to build risk assessment model. The Kaplan-Meier model has been applied to illustrate the rate of thrombus-free for the total patient population.Incidents of PICC-DVT was calculated according to the CPR scores categories in the elderly inpatients. P-values for trend for PICC-DVT in relation to the CPR scores was tested with the Jonckheere Trend Test. And evaluate the predictive value of the model with the receiver-operating characteristic curve (ROC curve).Results:1. According the standard, 564 cases were involved in this study, 532 cases of male and 32 females. The patients aged between 60 ? 102 years,mean age(88.13±7.26) years. The occurrence of PICC-DVT were 141 cases, the rate of thrombosis was 25%. The patients with PICC mean day 149.50± 117.71 days, the median time to thrombosis was 35 ( 13.52?107.51) days.2. Univariate analysis showed that :senile dementia, atrial fibrillation,atherosclerosis,recent surgery, prior VTE, PICC or CVC catheter puncture history,Upper limb function limits are statistically significant (P<0.05) .Multivariate analysis showed that:atrial fibrillation(HR1.82,95%CI 1.279 to 2.576),atherosclerosis,(HR1.84,95%CI1.264 to 2.679), recent surgery(HR2.17,95%CI 1.216 to 3.874), prior VTE(HR3.68, 95%Ci 2.552to5.300) are independent risk factors for elderly inpatients.According to these multivariate analysis results,the risk assessment model was as following: PICC-DVT Riskindex=1.302?1+0.775x2+0.596?3+0.610?4 (?1: priorVTE;?2: recent surgery; ?3: AF; ?4:atherosclerosis) .X was assigned a value of 0 or 1 according to the exist the risk factors.3. All patients were divided into three groups: score=0?1 point was low-risk group, score =1?2 points was medium-risk group, score range >2 points was high-risk group.By trend test, with the increase of risk score, the risk of PICC-DVT increased significantly. Jonckheere analysis showed:compared with low-risk group,the medium-risk group occurred PICC-DVT to 2.1 times (95%CI [1.08-4.09],P=0.029) , high-risk group occurred PICC-DVT to 6.7 times(95%CI[3.94-11.64], P<0.001). Area under the Roc curve(AUC)of the risk assessment model was 0.73(95%CI 0.69-0.78).Conclusion:The elderly inpatients are a high risk group of PICC-DVT. Multivariate logistic regression analysis shows that atrial fibrillation, atherosclerosis, recent surgery, prior VTE are independent risk factors for elderly inpatients. The risk assessment model of PICC-DVT Risk index=1.302?1+0.775x2+0.596?3+0.610?4predicts PICC-DVT of elderly inpatients more accurate. |
PDF Full Text Request