Predictive Value Of Fragmented QRS On Nonresponse To Cardiac Resynchronization Therapy

Objective: To evaluate the relationship between fragmented QRS complexes(f QRS) on a surface 12-lead electrocardiography(ECG) and improvement of New York Heart Association(NYHA) class and left ventricular ejection fraction(LVEF) after cardiac resynchronization therapy(CRT) in patients with chronic congestive heart failure(CHF); To explore the relationship between number of leads with f QRS and improvement of NYHA class and LVEF after CRT; To discuss the value of f QRS in predicting the incidence of non-response to CRT.Methods: One hundred and fifty patients were enrolled in this study. Study populations were divided into f QRS group and non-f QRS group according to the existence of f QRS on ECG. Patients in f QRS group were divided into subgroup f QRS≧3 and subgroup f QRS<3 according to whether QRS fragmentation were present in ≧3 leads. All patients were receiving optimal medical therapy for CHF. All patients underwent standard atrioventricular(AV) and(VV) interventricular delay optimization guided by Doppler echocardiogram. Before CRT implantation as well as six months after implantation, NYHA class, ECG and echocardiography were evaluated. Response to CRT was defined as improvement of at least one NYHA class or improvement of LVEF≥5% 6 months after implanation. Patients were defined as nonresponders if they did not reach the prespecified standard or died of cardiac causes.The univariate and multivariate Logistic regression analysis were performed to evaluate the predictive value of f QRS on nonresponse to CRT.Results:(1)There were 112 patients in f QRS group and 38 patients in nonf QRS group.The improvement of NYHA class in non-f QRS group was quite obvious(3.33±0.47 vs 2.47±0.60 P<0.01) and there was obvious increase of LVEF(27.76±5.26% vs 34.78±7.04% P<0.01); There was no evident improvement of NYHA class in f QRS group(3.42±0.50 vs 3.21±0.62 P =0.06) and there was no obvious increase of LVEF(29.26±4.91% vs 31.56±8.21% P =0.08). Nonresponse rate of f QRS group was 52.63% and non-f QRS group was 18.75%. The variation of nonresponse rate between the two groups was statistically significant(P<0.01).(2) There were 17 patients in subgroup f-QRS ≧3 and 21 patients in subgroup f QRS <3.Analysis of the two subgroups showed that the improvement of NYHA class in subgroup f QRS≧3 were worse than in subgroup f QRS <3, while the improvement of LVEF in the two subgroups was similar.(3) Univariate Logistic regression analysis showed that f QRS was an independent predictor of nonresponse to CRT(OR=2.64, 95%CI, 1.26-5.54, P <0.05). In multivariate Logistic regression analysis, f QRS was still significantly related with CRT nonresponse after adjusted for other risk factors(OR=2.55,95%CI,1.07-6.06, P <0.01).Conclusions:(1) Patients with f QRS on baseline ECG have a higher chance of nonresponse to CRT than those without f QRS.(2) Patients with f QRS ≧3 leads on ECG have a poorer responsiveness to CRT than those with f QRS <3 leads.(3) FQRS is valuable in predicting the incidence of non-response to CRT.