【作者】 巩雪；

【导师】 舒先红； 宿燕岗； 潘翠珍；

【作者基本信息】 复旦大学 ， 内科学， 2010， 博士

【摘要】 第一部分右室流出道间隔部和右室心尖部起搏对左室同步性和心功能影响的临床研究目的：比较右心室流出道(RVOT)间隔部和右室心尖部(RVA)起搏对心功能正常患者的左室同步性、心功能和左室重构的影响。方法：96例高度或Ⅲ度房室传导阻滞的患者随机分为RVOT间隔部起搏组(A组,n=48)和RVA起搏组(B组,n=48),于术前及术后12个月进行二维及组织多普勒超声心动图检查,评价左室容积、左室收缩、舒张同步性及收缩、舒张功能等情况。采集二维超声心动图心尖四腔心切面测量舒张末左室容积(LVEDV).收缩末左室容积(LVESV),用Simpson法计算左室射血分数(LVEF).采集组织多普勒图像(TDI)连续5个心动周期进行脱机分析,测量左室12节段心肌收缩达峰时间(Ts),舒张早期达峰时间(Te),并计算12节段Ts标准差(Ts-12SD).最大差(Ts-Dif)和12节段Te标准差(Te-12SD).最大差(Te-Dif).同时测量左室12节段平均心肌收缩速度(Sm)和平均心肌早期舒张速度(Em)。结果：随访12个月,其中A组有2人失访,B组有4人失访。术前两组患者的基线情况及各超声指标均无差异(P均>0.05)。起搏治疗12个月后,B组的Ts-12SD(26.43±14.39 ms vs 35.53±15.53 ms,P＜0.05)和Te-12SD(23.82±8.90msvs 27.98±11.07ms,P<0.05)均较术前明显增加,A组无明显变化(Ts-12SD：24.98±15.49 ms vs 28.30±15.09 ms；Te-12SD:25.84±14.69 ms vs 28.31±10.12ms；P均>0.05)。A、B两组的Ts-Dif均较术前增加(P均<0.05),而Te-Dif无明显变化(P均>0.05)。术后1年A组的Ts-12SD和Ts-Dif比B组小(p<0.05),Te-12SD和Te-Dif两组间无差异(P>0.05)。起搏治疗12个月后,两组的Sm.Em均较术前降低,Sm(A组：5.03±1.14 cm/s vs 4.44±1.08 cm/s;B组：5.10±1.56cm/s vs 4.24±1.15 cm/s；P均<0.05),Em(A组：5.79±2.62 cm/s vs 4.43±1.65 cm/s:B组：5.42±1.95 cm/s vs 4.44±1.74 cm/s:P均<0.05),但两组的LVEF与术前相比无差异(P均>0.05)。术后12个月两组的LVEF.Sm和Em均无差异(P均>0.05)。起搏治疗12个月后,两组的LVEDV和LVESV与术前比无明显变化(P均>0.05),术后1年两组间比较亦无差异(P>0.05)。结论：RVOT间隔部起搏相对RVA起搏可产生更加同步的左室收缩,但对心功能正常的患者,起搏12个月并未使RVOT间隔部起搏在保护心功能、阻止左室重构等方面优于RVA起搏。第二部分常用临床和超声指标对心脏再同步化治疗(CRT)疗效的预测价值目的：比较和评价常用超声指标、血清氨基末端脑钠素前体(NT-ProBNP)及QRS波时限在预测CRT疗效中的价值,探寻预测CRT疗效的最佳参数。方法：选择2006年9月至2009年12月在我院因心衰住院行CRT治疗的患者80例,在CRT术前和术后6个月进行超声心动图检查,采集二维超声心动图测量左室舒张末内径(LVDD)、左室收缩末内径(LVDS)、左房内径(LAD)、左室舒张末容积(LVEDV)、左室收缩末容积(LVESV),用Simpson法计算左室射血分数(LVEF)。通过M型超声心动图获取室间隔与左室后壁间的收缩延迟时间(SPWMD)。采集脉冲多普勒图像计算主动脉射血前时间(APEI)、肺动脉射血前时间(PPEI),并计算室间电机械延迟(IVMD)。采集组织多普勒图像(TDI)进行脱机分析,测量左室12节段收缩达峰时间(Ts)并计算12节段Ts标准差(Ts-12SD)。于术前及术后7天检测患者血浆NT-ProBNP浓度并转换为对数值log (NT-ProBNP),同时测量术前及术后7天QRS波时限(QRSd),计算两者的差值(△QRSd)。以治疗后6个月,左室收缩末容积较术前减小≥15%为治疗有效,以此分为CRT有效组和无效组,比较两组比较上述各指标术前术后变化,探讨其对CRT疗效的预测价值。结果：共有70例患者完成6个月随访。CRT有效率61%(43/70)。CRT有效组和无效组术前的LVDD、LVDS、LAD、LVEDV、LVESV和LVEF均无统计学差异(P均>0.05)。术后6个月CRT无效组的上述指标与术前比较无显著改变(P均>0.05),CRT有效组的LVDD、LVDS、LAD、LVEDV和LVESV较术前显著减小(p均< 0.05), LVEF明显改善(P<0.05)。CRT有效组和无效组术前的SPWMD、APEI、PPEI和IVMD无统计学差异(p均>0.05),CRT有效组的术前Ts-12SD高于无效组(42.03±16.33 ms vs 33.04±8.33 ms,P=0.018)。术后6个月CRT有效组的Ts-12SD较术前减小(42.03±16.33 ms vs 29.40±8.96 ms,p<0.05),余指标与术前比较无统计学差异,而CRT无效组的上述各指标较术前均无统计学改变(P均>0.05)。术前Ts-12SD对CRT疗效有预测价值(AUC=0.703,P=0.019),当Ts-12SD取>34.65ms时敏感度及特异度最大,敏感度61.8%,特异度70.6%。术前log (NT-ProBNP)水平对CRT疗效有预测价值(AUC=0.75,P=0.005),当log (NT-ProBNP)取≤3.04(即NT-ProBNP取≤1096pg/ml)时,敏感度及特异度最大,敏感度48.3%,特异度94.1%。术前QRSd、△QRSd及其他同步性指标均不能预测CRT疗效(p均>0.05)。结论：在超声心动图评价同步性的常用指标中,Ts-12SD尚能预测CRT疗效,但其预测价值尚不满意,有必要寻找更可靠的预测CRT疗效的超声新指标。术前NT-ProBNP水平对CRT疗效有一定的预测价值。术前QRSd及△QRSd不能预测CRT疗效。第三部分心脏再同步化治疗(CRT)对左室同步性和心功能影响的实验研究3.1超声新技术评价及预测心脏再同步化治疗(CRT)疗效的实验研究目的：通过建立快速起搏诱发心衰及CRT纠治心衰的动物模型,探讨超声新指标在CRT中的应用价值。方法：8-11个月的比格犬21只随机分为A组(CRT组,n=10),B组(非CRT组,n=7)和C组(假手术组,n=4)三组,三组均植入左室心外膜电极、右心室和右心房心内膜电极。A、B组以VOO模式260次/分快速起搏右心室诱发心衰,随后A组更换为CRT起搏器起搏4周,B组不予治疗自由恢复,C组不启动起搏。分别在术前、心衰后、CRT或自行恢复2周和4周以及起搏电极植入3周(C组)行二维、三维超声心动图检查,获取如下同步性指标：组织多普勒分析左室12节段达最大收缩速度时间的标准差(Ts-12SD)、斑点追踪显像分析左室乳头肌水平6节段QRS波起点至径向应变、圆周应变分别达峰时间的标准差(Trs-6SD、Tcs-6SD),实时三维超声心动图分析左室16节段达最小收缩容积时间的标准差(Tmsv-16SD),并测量左室射血分数(LVEF)、左室舒张末容积(LVEDV)及收缩末容积(LVESV)。以CRT治疗后4周,左室舒张末容积较心衰时减小≥15%为治疗有效,比较上述各同步性指标术前术后变化,并探讨其对CRT疗效的预测价值。结果：A组有2只,B组有1只比格犬因严重心衰死亡。手术成功率82.4%(14/17)。快速起搏2周后,A、B两组的LVEF较术前降低(58.67±3.26%vs31.97±1.84%,P<0.05), LVEDV (26.53±3.17 ml vs 35.50±6.46 ml, P<0.05)和LVESV (9.61±2.01ml vs 24.32±4.60 ml,P＜0.05)较术前增加,Ts-12SD(1.98±0.88%vs 5.47±2.28%,P<0.05)、Trs-6SD(1.70±0.78%vs 1 1.12±7.53%,p<0.05)、Tcs-6SD(3.50±0.89%vs 9.13±1.65%,P<0.05)以及Tmsv-16SD(0.94±0.24% vs 3.70±0.82%,P＜0.05)亦较术前明显增加,左室内各节段收缩呈现不同步,心电图QRS波时限较基线增宽(57.62±13.00 ms vs81.18±20.27 ms,P<0.05)。A组在CRT治疗4周后,LVEF.LVEDV.LVESV及同步性指标均有显著改善(p均<0.05),B组在关闭起搏器4周后,LVEF. LVESV及同步性指标较心衰时亦明显恢复(P均<0.05),但LVEDV无改善(34.6±5.1 ml vs 33.8±3.3 ml.P＞0.05).A组CRT4周与B组自由恢复4周相比,LVEF (57.0±2.4% vs 47.5±7.5%,P＜0.05).LVESV(14.3±2.2 ml vs 17.9±2.2 ml,P<0.05)、TS-1 2SD(2.26±1.21%vs 3.61±0.80%,P<0.05)、Trs-6SD(4.04±1.58%vs 6.83±3.15%,p<0.05)、Tcs-6SD(6.10±1.70%vs7.80±1.22%,p<0.05)和Tmsv-16SD(1.95±0.64% vs 3.29±0.93%,P＜0.05)差异亦有统计学意义,A组优于B组。相关性分析表明Ts-12SD.Tmsv-16SD.Trs-6SD.Tcs.6SD与LVEF呈负相关(r分别为-0.77、-0.86、-0.75、-0.83,p均<0.01),与LVEDV呈正相关(r分别为0.44、0.66、0.66、0.73,P均<0.01)。Trs-6SD对CRT疗效最有预测前景(AUC=0.85,p=0.054)。当Trs-6SD取≥12.16%时敏感性和特异性最大,分别为83.3%和100%。结论：VOO模式260次/分起搏2周可以成功建立犬心衰模型,且通过实时三维超声心动图和斑点追踪显像能有效检测出模型中存在的左室内不同步,CRT治疗后左室内不同步改善。在这些超声新指标中Trs-6SD对CRT疗效的预测价值最大,有望成为评价CRT疗效的可靠新指标。3.2心脏再同步化治疗(CRT)的可能分子机制探讨目的：通过建立快速起搏诱发心衰及CRT纠治心衰的动物模型,探寻CRT治疗心衰的可能分子机制。方法：8-11个月的比格犬21只随机分为A组(CRT组,n=10),B组(非CRT组,n=7)和C组(假手术组,n=4)三组,三组均植入左室心外膜电极、右心室和右心房心内膜电极,A、B组以VOO模式260次/分快速起搏右心室诱发心衰,随后A组更换为CRT起搏器起搏4周,B组不予治疗自由恢复,C组不启动起搏。于随访终点处死比格犬留取心肌标本,分别对比格犬左室心肌组织进行光镜、电镜检查,观察心肌细胞及线粒体形态、结构的变化。以GAPDH为内参,运用RT-PCR方法对线粒体乙醛脱氢酶2(ALDH2)及细胞钙转运相关蛋白-受磷蛋白(PLN)和肌浆网钙离子ATP酶(SERCA2a)等基因的转录水平进行检测,了解CRT治疗过程中有无分子生物学的变化。结果：A组有2只,B组有1只比格犬因严重心衰死亡。对这3只心衰比格犬左室心肌进行HE染色,可见心衰时小血管充血,肌丝较正常时变细,排列较紊乱。相对B组,A组经过4周CRT治疗后肌丝增粗明显。电镜下可见犬心衰时心肌细胞及线粒体均明显水肿,线粒体间、肌丝间的间隙增宽,成纤维细胞及溶酶体增多。相对B组,A组经CRT治疗4周后细胞、线粒体水肿恢复明显。随访终点时,根据电泳图像可见B组的PLN、SERCA2a及ALDH2灰度较A组及C组均不同程度减弱,ALDH2的表达在A、B两组间差异有统计学意义(P=0.028)。PLN、SERCA2a三组间无统计学差异(p>0.05)。结论：在快速起搏诱发心衰及CRT纠治心衰的模型中,CRT改善心功能、逆转左室重构的过程伴随着基因水平的改变,CRT有可能是通过增加线粒体ALDH2表达,改善线粒体能量代谢来发挥作用的。更多还原

【Abstract】 Part One:The Impact of Septal Right Ventricular Outflow TractPacing and Right Ventricular Apex Pacing on Left Ventricular Synchrony and Function in Patients with Normal Cardiac FunctionObjective:To compare the impact of septal right ventricular outflow tract (RVOT) pacing and right ventricular apex (RVA) pacing on left ventricular synchrony, cardiac function and remodeling in patients with normal cardiac function.Methods: A total of 96 consecutive patients with high or third degree atrial-ventricular block were enrolled and randomized into two groups:RVOT pacing group (n=48) and RVA pacing group (n=48). Tissue Doppler Imaging (TDI) and two-dimensional echocardiography (2DE) were performed to investigate left ventricular (LV) systolic and diastolic synchrony, LV systolic and diastolic function and LV volumes before pacemaker implantation and after 12 months of ventricular pacing. Using the 2DE apical 4 chamber views, we acquired the left ventricular (LV) end-diastolic volume (LVEDV), end-systolic volume (LVESV) and LV ejection fraction (LVEF) measured by Simpson’s rule. From color TDI, the time to peak myocardial systolic velocity during the ejection phase (Ts) and the time to peak myocardial early diastolic velocity (Te) were measured with referenced to QRS complex, the peak myocardial systolic velocities (Sm) and peak myocardial early diastolic velocities (Em) were recorded. The standard deviation of Ts (Ts-12SD) of all the 12 LV segments, and the maximal difference in Ts (Ts-Dif) between any 2 of the 12 LV segments were calculated. The standard deviation of Te (Te-12SD) of all the 12 LV segments and the maximal difference in Te (Te-Dif) were measured.Results:Two patients in group A and 4 patients in group B were lost during the 12 months of follow-up. There were no significant differences in baseline characteristics and all echocardiography parameters between the two groups. After 12 months of ventricular pacing, Ts-12SD (26.43±14.39 ms vs 35.53±15.53 ms, P<0.05) and Te-12SD (23.82±8.90 ms vs 27.98±11.07 ms, P<0.05) increased in group B, and remained unchanged in group A (Ts-12SD:24.98±15.49 ms vs 28.30±15.09 ms; Te-12SD:25.84±14.69 ms vs 28.31±10.12 ms; both NS). Ts-Dif prolonged (P< 0.05) and Te-Dif did not change (both NS) in both groups. Ts-12SD and Ts-Dif in group B were greater than those in group A (both P<0.05), while Te-12SD and Te-Dif were not different between the two groups at 12 months of follow-up (NS). After 12 months of ventricular pacing, Sm (group A:5.03±1.14 vs 4.44±1.08 cm/s; group B:5.10±1.56 vs 4.24±1.15 cm/s; both P<0.05) and Em (group A:5.79±2.62 vs 4.43±1.65 cm/s; group B:5.42±1.95 vs 4.44±1.74 cm/s; both P<0.05) decreased but LVEF did not change (NS) in both groups. LVEDV and LVESV did not change both in group A and group B (all NS). There were no significant differences with respect to Sm, Em, LVEF, LVEDV and LVESV in the two groups at 12 months of follow-up (all NS).Conclusions:Although RVOT septal pacing caused more synchronous LV contraction compared with RVA pacing, it did not benefit over RVA pacing in the aspect of preventing cardiac remodeling and preserving LV systolic function after 12 months of pacing in patients with normal cardiac function.Part Two:The Predictive Value of Clinical and Echocardiographic Parameters on Cardiac Resynchronization TherapyObjective:To compare and assess the value of echocardiographic parameters of M-mode, pulsed-wave Doppler and tissue Doppler imaging which were usually used for selecting cardiac resynchronization therapy (CRT) candidates, N-terminal pro-brain natriuretic peptide (NT-ProBNP), and the duration of QRS wave in predicting the response to CRT and to evaluate the possible best parameter to predict the response to CRT.Methods:Eighty patients accepted CRT because of refractory heart failure were enrolled from 2006.9 to 2009.12. Two-dimensional echocardiography (2DE), M-mode, pulsed-wave Doppler and tissue Doppler imaging (TDI) were performed before and after 6 months of CRT. Using the 2DE, we acquired the left ventricular (LV) diastolic diameter (LVDD), LV systolic diameter (LVDS), left atria diameter (LAD), LV end diastolic volume (LVEDV), LV end systolic volume (LVESV) and LV ejection fraction (LVEF) measured by Simpson’s rule. M-mode echocardiography, pulsed-wave Doppler and TDI were used to evaluate inter- and intra- ventricular dyssynchrony. M-mode echocardiography was used to acquire the septal- to- posterior wall motion delay (SPWMD), pulsed-wave Doppler was used to acquire aortic wave pre- ejection interval (APEI) and pulmonary wave pre- ejection interval (PPEI), and inter-ventricular mechanical delay (IVMD) was calculated. TDI were used to get the standard deviation of time to peak myocardial systolic velocity (Ts-12SD). Meanwhile, the concentration of NT-ProBNP was detected before and after 7 days of CRT implantation and transformed to logarithm log (NT-ProBNP). The duration of QRS wave (QRSd) were measured preoperation and after one week of CRT, and the difference of two QRS wave (AQRSd) was computed. At least 15% reduction in LVESV at the 6-month after CRT was defined as responders. We aimed to compare the changes of all parameters between responders and non-responders and to investigate the predictive value of ail the parameters.Results:Seventy patients complete the 6 months of follow up, and there were 43 (61%) responders. There were no significance between responders and non-responders in terms of LVDD, LVDS, LAD, LVEDV, LVESV, LVEF, SPWMD, PPEI, APEI and IVMD before CRT implantation (all NS), but responders had relatively high Ts-12SD than non-responders (42.03±16.33 ms vs 33.04±8.33 ms, P=0.018). After 6 months of CRT, the LVDD, LVDS, LAD, LVEDV and LVESV decreased significantly (P<0.05), LVEF increased significantly (P<0.05) and Ts-12SD decreased significantly (42.03±16.33 ms vs 29.40±8.96 ms, P<0.05) in responders, while there were no significant changes with respect to all the parameters in non-responders (all NS). Ts-12SD is relatively valuable in predicting CRT responders (AUC=0.703, P=0.019), Ts-12SD≥34.65ms has the relatively high sensitivity of 61.8% and specificity of 70.6%。The level of log (NT-ProBNP) preoperation could predict the response to CRT (AUC=0.751, P=0.005), log (NT-ProBNP)≤3.04 (NT-ProBNP<1096pg/ml) has the sensitivity of 48.3%, specificity of 94.1% to predict the response to CRT. The QRSd,△QRSd and other synchrony parameters could not predict the response to CRT (NS).Conclusions:Tissue Doppler imaging is relatively better than M-mode echocardiography and pulsed Doppler parameters, but it is still not satisfactory. It is necessary to look for more reliable echocardiography parameters to predict the response to CRT. The level of NT-ProBNP before CRT implantation is valuable in predicting the response to CRT. The QRSd and AQRSd could not predict the response to CRT. Part Three:The Animal Study of the Impact of Cardiac Resynchronization Therapy on Left Ventricular Synchrony and Cardiac Function3.1:The Value of New Echocardiography Parameters in Evaluating and Predicting the Response to Cardiac Resynchronization Therapy: An Animal studyObjective:To establish the animal model in terms of rapid pacing induced heart failure and possible mechanism of cardiac resynchronization therapy (CRT) in treating heart failure, and to investigate the value of new echocardiography parameters in appraising the therapeutic effect of CRT.Methods:Twenty-one adult Beagle dogs were divided into Group A (CRT group, n= 10), Group B (non-CRT group, n=7) and Group C (control group, n=4). All of them were implanted left ventricle epicardial electrodes, right atrium and right ventricle leads. Group A and’B received rapid right ventricular pacing (VOO mode,260bpm) to induce heart failure. When their LVEF decreased below 35%, dogs in Group A treated with CRT and dogs in Group B were just terminated pacing. Dogs in Group C were not received rapid pacing and used as control group. Echocardiography including Tissue Doppler Imaging (TDI), Real-time three-dimensional echocardiography (RT3DE) and Speckle tracking strain imaging (STSI) and electrocardiography were performed at baseline, heart failure,2 and 4 weeks after CRT or termination of pacing. The dyssynchrony parameters included the standard deviation of time to maximum systolic velocity of the 12 LV segments (Ts-12SD) by TDI, the standard deviation of time to minimum systolic volume of the 16 LV segments (Tmsv-16SD) by RT3DE, standard deviation of time from onset of QRS to peak radial strain (Trs-6SD) and peak circumferential strain (Tcs-6SD) of the 6 middle LV segments by STSI. LV ejection fraction (LVEF), LV end diastolic volume (LVEDV) and LV end systolic volume (LVESV) were measured by RT3DE. At least 15% reduction in LVEDV at the 4-weeks after CRT was defined as responders. We aimed to compare the changes of all parameters in the animal model and to investigate the predictive value of all the parameters.Results:Two dogs in Group A and one dog in Group B died of severe heart failure. The success rate was 82.4%(14/17). In Group A and B,2 weeks of rapid right ventricular pacing decreased LVEF (58.67±3.26% vs 31.97±1.84%, P<0.05), enlarged LVEDV (26.53±3.17ml vs 35.50±6.46 ml, P<0.05) and LVESV (9.61± 2.01 ml vs 24.32±4.60 ml, P< 0.05), increased Ts-12SD (1.98±0.88% vs 5.47±2.28%,P＜0.05)、Trs-6SD (1.70±0.78% vs 11.12±7.53%, P<0.05)、Tcs-6SD (3.50±0.89% vs 9.13±1.65%, P<0.05) and Tmsv-16SD (0.94±0.24% vs 3.70±0.82%, P<0.05), dyssynchronized LV segments and prolonged QRS duration (57.62±13.00 ms vs 81.18±20.27 ms, P<0.05). After 4 weeks of CRT, the LVEF, LVEDV, LVESV, Ts-12SD, Tmsv-16SD, Trs-6SD and Tcs-6SD of Group A were all improved (P<0.05). After 4 weeks of pacing termination, LVEF, LVESV and the dyssynchronized index of Group B also recovered significantly (P<0.05), but LVEDV remained enlarged (34.6±5.1 ml vs 33.8±3.3 ml, NS). Covariance analysis showed that LVEF (57.0±2.4% vs 47.5±7.5%, P<0.05), LVESV (14.3±2.2 ml vs 17.9±2.2 ml, P<0.05), Ts-12SD (2.26±1.21% vs 3.61±0.80%, P<0.05), Trs-6SD (4.04±1.58% vs 6.83±3.15%, P<0.05), Tcs-6SD (6.10±1.70% vs 7.80±1.22%, P<0.05) and Tmsv-16SD (1.95±0.64% vs 3.29±0.93%, P<0.05) in Group A were better than in Group B. Ts-12SD, Tmsv-16SD, Trs-6SD and Tcs-6SD had good negative correlation with LVEF (r was-0.77、-0.86、-0.75、-0.83 relatively, all P<0.01) and positive correlation with LVEDV (r was 0.44、0.66、0.66、0.73 relatively, all P<0.01). Trs-6SD is relatively valuale in predicting the response to CRT (AUC=0.85, P=0.054), Trs-6SD≥12.16% has the relatively high sensitivity of 83.3% and specificity of 100%。Conclusions:Rapid right ventricular pacing in 260 bpm could induce obvious heart failure, and in the canine model of CRT after rapid right ventricular pacing induced heart failure, RT3DE and STS1 can effectively detect left ventricular dyssynchrony. CRT improved cardiac dyssynchrony. Among the new echocardiography parameters, Trs-6SD might be an optimal parameter to predict the effect of CRT.3.2: The possible molecular mechanism of cardiac resynchronization therapyObjective: To study the possible molecular mechanism of cardiac resynchronization therapy (CRT) through establishing the animal model in terms of quick pacing induced heart failure and how CRT treat heart failure. Methods:Twenty-one adult beagle dogs were divided into Group A (CRT group, n= 10), Group B (non-CRT group, n=7) and Group C (control group, n=4). All of them were implanted left ventricle epicardial electrodes, right atrium and right ventricle leads. Group A and B received rapid right ventricular pacing (VOO mode,260bpm) to induce heart failure. When their LVEF decreased below 35%, dogs in Group A treated with CRT and dogs in Group B were just terminated pacing. Dogs in Group C were not received rapid pacing and used as control group. After 4 weeks of CRT or terminated pacing, LV myocardium was taken out for HE staining and elcectron microscope inspection. The mRNA transcription levels of contractile, calcium regulatory proteins such as phospholamban (PLN) and sarcoplasmic reticulum Ca (2+)-ATPase protein (SERCA), and mitochondria related proteins such as acetaldehyde dehydrogenase-2 (ALDH2) were measured by reverse transcription polymerase chain reaction (RT-PCR) and normalized for glycer-aldehyde 3-phosphate dehydrogenase (GAPDH).Results:Two dogs in Group A and one dog in Group B died because of severe heart failure. By HE staining of the heart failure myocardium, we observed small vessel congestion, myofilament thinningz, and myofilament of group A thickened obviously compared with group B. Electron microscope showed cellular edema, mitochondrial swelling and fine myofibril in heart failure cardiac muscle. After 4 weeks of CRT, there were no mitochondria swelling and on cellular edema in Group A, after terminated pacing for 4 weeks, there were still different extent of mitochondria swelling in Group B. The results of RT-PCR show that the grey level of PLN, SERCA2a and ALDH2 in Group B were all lower than in Group A, and expression of ALDH2 was significantly lower than in Group A (P=0.028), but there were no significant difference in PLN and SERCA2a among three groups (MS).Conclusions:In the canine model of CRT after rapid right ventricular pacing induced heart failure, functional improvement related to CRT is associated with reversal of the HF related gene programming, CRT may produce a marked effect through mitochondria ALDH2 and energy metabolism.更多还原