Font Size: a A A

Clinical Value Of Multivariate Analysis For Assessing The Risk Of Thromboembolism In Patients With Persistent Atrial Fibrillation

Posted on:2006-07-14Degree:MasterType:Thesis
Country:ChinaCandidate:G LiFull Text:PDF
GTID:2144360155969326Subject:Department of Cardiology
Abstract/Summary:PDF Full Text Request
Background:Thromboembolism is the most common complication of persistent atrial fibrillation (PAF). Anticoagulation treatment can effectively prevent it. But the side effect of hemorrhage and the problem of supervision brought by anticoagulation make the imminent need of effective measures to assess the risk of thromboembolism. Thrombosis is a procedure that involves many factors. Recently more and more evidents have proved that thrombosis has the relationship with not only abnormal local haemodynatics but also increase of thrombin, activation of platelet, damage of endothelium, deactivation of anticoagulation and fibrinolytic system. The use of transesophageal echocardiography(TEE)greatly improved the detection of left atrial thrombies,so it has become the main measure to assess risk of thromboembolism in AF patients. However,some studies suggested that TEE may miss the thrombies whose dimensions were under 2cm and the new ones. So it's necessary to detect thrombosis by studying the coagulation system. If combining TEE with thrombus molecular markers,the assessment mistakes can great decrease,which has the important role to guide anticoagulation treatment,reduce incidence of thromboembolism and hemorrhage.Objective: By detecting several sensitive thrombus melucular markers considering related clinical and echocardiographic factors, this study is to investigatethe clinical value of multivariate analysis for assessing the risk ofthromboembolism in patients with persistent atrial fibrillation.Patients and Methods:69 patients with PAF were divided into 2 groups: thromboembolism group (n=30) and non-thromboembolism group (n=39). 35 patiants with heart diseases but without atrial fibrillation were selected as control. All subjects underwent trans-thoraial echocardiography ( TTE) to detect LAdiameter, left ventricular ejective factor (LVEF) .48 AF patients underwent TEE to detect LAA area, LAA velocity and LASEC. The plasma TAT, GMP-140, vWF and D-D levels were assayed with ELISA,the plasma AT-III level was assayed with spectrophotometric assay. In the end,the multivariate analysis was made for the results.Results: (1) In the thromboembolism group(n=30), 12 patients were attacked by embolism(8 cerebral embolism patients , 4 peripheral embolism patients,all having typical symptom and being confirmed by CT or MRI); 18 thrombus patients were identified by TTE, TEE or surgery,22 thrombies(9 in LA,13 in LAA).The largest thrombi was 98mm × 70mm,the smallest one was 6mm × 2.2mm.(2) There was no significant difference in sex, AF duration and associated diseases(hypertention, coronary heart disease)among three groups; The age of the thromboembolism group was 66.3 ±11.9 years old,significantly older than the non-thromoembolism group(63.1±11.4 years old, P<0.05); There was 13 valvular diease patients in thromboembolism group,much more than the latter two groups(5 and 2 valvular disease patients respectively, P<0.01).(3) Left atrial dimension in thromboembolism group (44.8±6.8mm)was greatly larger than the latter two group(36.5±5.2mm,36.0 ± 6.9mm respectively,P<0.01); there was no significant difference in LVEF level among three groups(54.6±10.4%,\ 57.9+12.2% > 57.9+12.0% respectively,P>0.05).In TEE indexes, rate of LASEC (≥ 2 grade) was significantly increased, LAA filling and emptying velocities were significantly decreased in thromboembolism group(50%, 0.13 + 0.05m/sfP0.12 + 0.04m/s,respectively),compared with non-thromoembolism group(ll.l%,0.25 + 0.14m/s和 0.22+0.14m/s,respectively,P <0.01);LAA area was not the case( 1130.3 +343.3 mm2,1100.9±468.1 mm2,respectively,P>0.05);In thrombus melucular markers, plasma TAT,GMP-140,vWF,D-D levels in thromboembolism group were 19.4 + 10.8ng/ml, 14.0 + 5.7ng/ml, 156.2 ± 45.8%,0.82 ± 0.49 u g/ml,respectively,in non-thromboembolism group were 10.4 ± 9.7ng/ml, 11.4 ± 4.1ng/ml, 119.7 ± 44.0%,0.51±0.45u g/mLrespectively, in the control were 7.8 ± 4.8ng/ml, 9.1 + 5.7ng/ml, 96.6+34.3%, 0.27±0.20 u g/ml,respectively.There was greatly significant difference among three groups (P values were all less than 0.01) ; plasma AT-III level in thromboembolism(98.3+29.1%)was lower than the latter groups(108.1± 35.8%,108.3+30.4%,repectively),but have no significance (P>0.05).(4) In multivariate logistic regression, increase of LA diameter (P=0.012) > vWF (P=0.030)> D-D (P=0.003),LASEC^2grade(P=0.028) were independently associated with thromboembolism;the related risk degree of increase of D-D (OR= 3.281, 95%CI 0.819~13.150) was greater than increase of LA diameter (OR=1.387,95%CI 1.129—1.703) and vWF(OR= 1.031,95%CI 1.003—1.059), LASEC^2 grade (OR=0.112,95%CI 0.016—0.791 ).In addition,among the four risk factors,vWF had the liner relationship with LA dimension (r=0.804,P=0.014) .(5) Receiver Operating Characteristic Curve(ROC Curve)showed that^Area Under the Curve(AUC)of LA diameter was 0.822(95%CI 0.710~0.934),of vWF was 0.832(95%CI 0.704—0.961),all larger than 0.7,with P value all less than 0.01,so LA diameter and vWF had diagnostic value to some extent. Moreover,the AUC of D-D was 0.905(95%CI 0.740—0.971)>0.9,and P<0.01,so D-D had higher diagnostic value than others.When the right diagnostic index(r value) reached maxism,their cut points were 41.5mnK 150.5% ^ 0.9ug/ml respectively.whereares AUC of LASEC was 0.568(95%CI 0.479 — 0.767)<0.7,and P>0.05,so LASEC had no diagnostic value,whose cut point was 2 grade (++).(6) Among single risk factor, the sensitivity and negetive prognostic valueof LA diameter (69.2% and 72.4%) to detect thrombus was the highest, and the specificity and positive prognostic value of D-D(82.1% and 76.2%) was the highest; Along with two risk factors,the specificities were 92.9%,92.9%,96.4%,respectively, positive prognostic values were 85.7%,83.8%,91.7%,respectively, higher than thosewith single factor(75.0%,75.0%,82.1%, and 72.0%,70.8%,76.2%,respectively); the sensitivities were 46.2%,38.5%,42.3%,respectively, negetive prognostic values were 65.0%,61.9%,64.3%,respectively,lower than those with single factor; Along with three risk factors, the specificity(100%) and positive prognostic value(100%) were much higher,but the sensitivity(23.1%)and negetive prognostic value(58.3%) were much lower.Conclusions: (1) Increase of LA diameter vWF^ D-D, LASEC^2 grade are independent risk factors of thromboembolism in patients with RAF. The assessment value of increase of D-D is higher than other factors.Among the four risk factors,vWF has the liner relationship with LA dimension.(2) Among the independent risk factors, D-D has the highest value to indentify thromboembolism in patients with PAF, LA diameter and vWF follow it.(3) Analysing single risk factor has higher sensitivity,which can be used to detect thromboembolism in patients with PAF at the first time; Considering more risk factors, the specificity of detecting thrombus is higher and the risk assessment of thromboembolism in patients with PAF is more reliable.
Keywords/Search Tags:Persistent Atrial Fibrillation, Thromboembolism, Risk Factor, Multivariate Analysis
PDF Full Text Request
Related items