Feasibility of In-Vivo Simulation of Acute Hemodynamics in Human Atrial Fibrillation

Figures

• 

  Fig 1. General principle of the method for simulation of AF. (A) QRS complexes were identified in an Holter ECG recording of AF to obtain a text string of the corresponding RR intervals. (B) The sequence of RR intervals was loaded to an Arduino-based microcomputer. The microcomputer generated square-wave pulses according to the RR interval sequence. (C) The pulses from the microcomputer were sensed by a cardiac stimulator, which was set to a sense-pace mode. (D) Pacing was performed by catheters in the coronary sinsus (i.e., left atrial pacing) and at the His bundle region. A pigtail cathter was inserted antegradely to measure the LV pressure.

• 

  Fig 2. Characteristics of the template AF sequences. The figure shows histograms of the RR intervals and of the absolute differences of the subsequent RR intervals that were contained in the template AF sequences. It can be seen that both sequences contained a broad range of homogenously distributed RR intervals, which enabled to study the effects of heart rhythm irregularity. The dark bars represent mean cycle length (667 ms and 462 ms—i.e., 90/min and 130/min, respectively).

• 

  Fig 3. Sample of ECG with corresponding LV pressure signal during simulated AF. The figure shows a 10s sample of ECG and a corresponding LV pressure tracing recorded in all the study patients during two separate runs of the simulated AF. It can be appreciated that (i) the His pacing generated narrow QRS complexes thus indicating intrinsic activation of the ventricles, (ii) the RR intervals followed the same pattern in all the patients at both runs, (iii) the simulated AF generated similar LV pressure tracings among all the patients, (iv) the LV pressure tracings were almost identical within each patient at the two separate runs. Pt = patient.

• 

  Fig 4. Effect of the simulated AF on the LV performance. The colored lines represent average values in the individual patients during three different pacing modes (atrial regular [reg], simultaneous atrioventricular [AV], simulated AF [AF]) at two different heart rates (90/min and 130/min). The black line with the errorbars represents mean and standard errors. */**/*** = P-value <0.5/<0.01/<0.001 by paired t-test with Holm’s correction; n.s. = non-significant difference (P-value > 0.5);

• 

  Fig 5. Reproducibility of the LV hemodynamics during two separate runs of simulated AF. (A) The colored lines represent average values in the individual patients obtained at two separate runs of the simulated AF. The black line with the errorbars represent mean and standard errors of all the patients. The p values were obtained by a paired t-test. (B) The graph shows excellent correlation between LV systolic pressure measured in the same patient during two separate runs of simulated AF. The data points were pooled from all the patients.