Analysis Of Influencing Factors Of Acute Radiation-induced Heart Disease

Objective：To study acute radiation-induced heart disease (RIHD) inpatients with thoracic tumor radiation therapy, in order to explore the effect ofbasic cardiac functional status, clinical factors and cardiac exposure doses orvolume on acute RIHD. To analysis the effect of whole heart and left ventricleexposure doses-volume parameters on cardiac function after radiationtreatment, and To compare RTOG acute radiation morbidty scoring schemaand Common Terminology Criteria for Adverse Events,3.0version(CTCAE3.0) on acute radiation-induced heart damage(RIHD).Methods: From December2008to August2012,109patients with chesttumor received conformal and intensity-modulated radiotherapy in the fourthhospital of Hebei Medical University were selected. There were33female,76male.The median age was61years, and57patients were lung cancer,52wereesophageal cancer. They were scanned with3mm layer spacing CT, and CTimages were transmitted to Pinnacle7.6c or8c planning systems, thenthree-dimensional reconstruction occurred. Target dose distribution andheart exposure doses were analyzed by three-dimensional treatment planningsystems. Irradiation was delivered at1.8～2.2Gy,5times per week to atotal dose of50～66Gy. Before radiotherapy, after radiotherapy and within3months since the irradiation began， cardiac enzymes, troponin I,echocardiography chart, ECG and chest CT were detected, and then thegrading standards of CTCAE3.0and RTOG on RIHD were evaluated, in orderto explore the effect of basic cardiac functional status, left ventricle exposuredoses-volume parameters on acute RIHD. Compare acute RIHD of RTOG andCTCAE3.0conformations. SPSS13.0software package was used for statisticalanalysis. t test for normal distribution data of two sample comparisons, Mann-Whitney rank sum test for non-normally distributed data,2test foranalysis of variance for multiple samples, Spearman method for relevantvariables analysis, logistic regression and univariate analysis for clinicalfactors. Variables were analyzed with Logistic forward stepwise regressionand multivariate analysis, McNemar-Bowker test was used to compare thecardiac injury grading and kappa value Correlation analysis were calculated tojudge the conformation of the two cardiac injury grading standards.Results:1According to CTCAE3.0,87cases with acute RIHD occurred duringthe first three months, the total incidence rate was79.8%(87/109),73caseswere1grade,14cases were2grade, and3grade or above level of acuteRIHD was not found. Center effusion was5cases (4.6%), electrocardiogramabnormal was54cases (49.5%), troponin I increased was14cases (12.8%),myocardial enzymes elevated was26cases (23.9%), left ventricular systolicfunction decreased was24cases (22.0%), left ventricular diastolic functionreduced was15cases (13.8%), valve morphology, function changed was24cases (22.0%).2Patients with various clinical factors such as age, gender, KPS scorestatus, smoking history, smoking index, history of heart disease, history ofsurgery, chemotherapy, TNM stage, or weight loss did not show any effect onacute RIHD (P>0.05).3The basic heart function in patients with acute RIHD and in patientswith no acute RIHD showed no significant difference (P>0.05).4Heart dose: Dmax was6126.4cGy±1516.0cGy, the MHD was2130.5cGy±1279.6cGy, and V5, V10, V20, V30, V40, V50, V60were64.6%±32.9%,56.4%±33.6%,43.3%±30.5%,31.6%±24.4%,21.7%±18.8%,12.4%±11.5%,5.5%±6.5%, D5, D50, D100were4945.0cGy±1890.7cGy,1720.6cGy±1456.8cGy,282.0cGy±371.5cGy， Heart NTCP was10.7±17.6.5V60, NTCP in patients with acute RIHD were higher than those with noacute RIHD, the difference was statistically significant (t=2.140, P=0.032, t =2.088, P=0.037). In order to find the appropriate threshold, ROC curveanalysis was used for V60and NTCP. The results showed that the area underthe curve was0.747and0.675, the accuracy of prediction was up to mediumand close to medium, boundary values were5.04%and1.5.6Physical factors associated with acute RIHD was involved in thelogistic regression model, it was showed that the whole heart V60was aindependent risk factor for the occurrence of acute RIHD(P=0.012). Theincidence of acute RIHD in patients with V60≥5.04%was93.3%, which wassignificantly higher than that of patients with V60<5.04%, whose incidencewas70.3%(2=8.692, P=0.003).7Before radiotherapy, after radiotherapy, and radiotherapy within3months since the irradiation began, Anteroposterior diameter of the left atrium,aortic flow velocity, and E/A had significant difference. Compare with thosebefore radiotherapy, Anteroposterior diameter of the left atrium, aortic flowvelocity, and E/A after or radiotherapy within3months since the irradiationbegan were both induced（P＜0.05）. Compare with those after radiotherapy,Anteroposterior diameter of the left atrium, aortic flow velocity, and E/A wereboth increased（P＜0.05）.8According to the CTCAE3.0, the incidence of acute RIHD was79.8%,which showed significant difference than that of RTOG standard (2=32.000，P=0.000).which had a middle consistency with RTOG standard（Kappa=0.417，U=6.351,P=0.000）,which was correlated which RTOGstandard（Pearson，sR=0.63，P=0.000）.9Left ventricular volume parameters: left ventricular maximum dose was4511cGy±2252.4cGy, the average dose of left ventricular was1487.6cGy±1251.7cGy, V5, V10, V20，V30, V40, V50, V55, V60were48.3%±40.5%,45.7%±37.6%,30.2%±31.6%,17.7%±21.5%,10.1%±16.4%,5.2%±12%,3.6%±10.2%,2.3%±8.6%, D5, D50, D100were3158.3cGy±2058.6cGy,1256.7cGy±1320.3cGy,326.6cGy±530.1cGy.10Results showed no relation between left ventricular volume-doseparameters and acute RIHD. V50, V55in patients with acute RIHD were higher than those with no acute RIHD (P=0.026; P=0.026). In order to findthe appropriate threshold for indication of left ventricular injury, ROC curveanalysis was used, the results showed that the area under the curve was0.673and0.660, the accuracy of prediction is close to medium, boundary valueswere1.78%and0.48%.11Physical factors which were associated with acute the radioactivity leftventricular diastolic function impairment were involved into the logisticregression model, the results showed that V50was an independent influencingfactors for left ventricular diastolic function of acute RIHD(P=0.025)。Theincidence of acute radioactive left ventricular diastolic function injury inpatients with V50≥1.78%was23.8%, significantly higher than7.9%ofpatients with V50<1.78%(2=4.998, P=0.025).Conclusions:1Clinical factors or basic cardiac functional status of patients was nosignificant correlation with acute RIHD.2V60and NTCP of heart are the main physical factors for acute RIHD,and V60is an independent factor.3Compare with those before radiotherapy, Anteroposterior diameter ofthe left atrium, aortic flow velocity, and E/A after radiotherapy were induced.aortic flow velocity within3months since the irradiation began Compare withthose after radiotherapy, Anteroposterior diameter of the left atrium, aorticflow velocity, and E/A within3months since the irradiation began were bothincreased.4The grading standards of acute RIHD in RTOG has significantdifferences with that in CTCAE3.0.5From beginning of radiotherapy to3months, E/A change have relatedto left ventricular dose-volume parameters, while left ventricular diastolicfunction parameters changes has no related to left ventricular dose-volumeparameters.6V50and V55of left ventricular are main physical indicator factors foracute left ventricular diastolic function impairment, and the former is also an independence indicator. If V50≥1.78%, acute radiation injury incidence ofleft ventricular diastolic function is significantly increased.