Reintroducing Heart Sounds for Early Detection of Acute Myocardial Ischemia in a Porcine Model – Correlation of Acoustic Cardiography With Gold Standard of Pressure-Volume Analysis

Figures

• 

  FIGURE 1. Experimental setup. (A) Schematic representation of the animal experiment. The swine is placed in supine position and instrumented with a series of different probes and catheters allowing multimodal realtime recordings. (B) Representative fluoroscopic images of myocardial infarction generation before (above) and after (below) balloon inflation. (C) Timeline of our study composed of three distinct phases: baseline, pre-ischemia (falling between coronary engagement and balloon inflation) and myocardial ischemia. Continuous recordings were performed through the pre-ischemia and myocardial ischemia for 5 min after coronary occlusion.

• 

  FIGURE 2. Conceived acoustic and pressure-volume derived parameters during baseline and after coronary occlusion. Acoustic systolic times were defined as follows: Q to S1 (electromechanical activation time or EMAT), S1 to S2 (left ventricular systolic time or LVST), and Q to S2 (electromechanical systolic time or EMST). Acoustic diastolic times were S2 to S1 (left ventricular diastolic time or LVDT), S2 to Q (left ventricular perfusion time or LVPT). Pressure-volume derived systolic times were: Q to +dP/dt (pre-ejection period), +dP/dt to –dP/dt (left ventricular systolic time) and Q to –dP/dt (electromechanical systolic time). The systolic time obtained by pressure-volume was delimited within –dP/dt to diastole’s end according to the company booklet. Upon occlusion a series of changes in shown tracings can be noticed: S1 and S2 intensities decrease, S3 and S4 start appearing, both maximal dP/dt and minimal dP/dt absolute values decrease indicating systolic and diastolic impairment, respectively. In addition, smaller excursions of left ventricular pressures and volumes (LVP and LVV) of tracings translate into lower systolic pressure generation and increased end-systolic volumes.

• 

  FIGURE 3. Scatter plots of overlapping interval times obtained by acoustic cardiography and pressure-volume analysis. (A) PV Heart Rate vs. AC Heart Rate; (B) Q to max dP/dt vs. QS1; (C) max to min dP/dt vs. S1 to S2 (LVST); (D) Q to min dP/dt vs. QS2; (E) Diastolic Filling Time vs. S2 to S1 (LVDT); (F) Diastolic Filling Time vs. S2 to Q (LVPT). Color key: blue (baseline), pre-ischemia (purple), red (myocardial ischemia). Each dot represents a 30-s assessment. Pressure-Volume data are reported on the x-axis, data obtained from acoustic cardiography are on the y-axis.

• 

  FIGURE 4. Scatter plots of systolic and diastolic performance obtained by acoustic cardiography and pressure-volume analysis. (A) Cardiac Output vs. S3 Display Value; (B) Cardiac Output vs. S4 Display Value; (C) Isovolumic Relaxation Costant (Tau) vs. S3 Display Value; (D) Isovolumic Relaxation Constant (Tau) vs. S4 Display Value. Color key: blue (baseline), pre-ischemia (purple), red (myocardial ischemia). Each dot represents a 30-s assessment. Pressure-Volume data are reported on the x-axis, whereas data obtained from acoustic cardiography are on the y-axis.

• 

  FIGURE 5. Heatmaps. (A) Reported fold changes obtained by comparing baseline measurement with their respective at each minute for 5 min upon coronary occlusion. (B) Reported two tailed t-test p-values obtained by with the same exact approach employed for the other heatmap. HR, heart rate; SV, stroke volume, CO, cardiac output, EF, ejection fraction; EDP, end diastolic pressure; EDV, end diastolic volume; ESP, end systolic pressure; ESV, end systolic volume; EDPVR, end diastolic pressure-volume relationship; ESPVR, end systolic pressure-volume relationship; DFT, diastolic filling time; PRSW, pre-recruitable stroke work; DT, diastolic time; PT, perfusion time; LVST, left ventricular systolic time; DV, display value. Significant changes are marked with an asterisk sign (∗).

• 

  FIGURE 6. Single Datapoint Charts AC parameters shown for each animal from baseline, pre-ischemia, and ischemia phases. Dashed lines delimit the three different experimental steps. (A) heart rate, (B) QS1, (C) QS2, (D) Diastolic time, (E) Perfusion time, (F) Left ventricular systolic time, (G) QS1 shown in percentage of a complete cardiac cycle, (H) QS2 shown in percentage of a complete cardiac cycle, (I) Diastolic time shown in percentage of a complete cardiac cycle, (J) Perfusion time shown in percentage of a complete cardiac cycle, (K) Left ventricular systolic time shown in percentage of a complete cardiac cycle, (L) S1 intensity, (M) S2 intensity, (N) S3 intensity, (O) S4 intensity, (P) S3 display value, (Q) S4 display value. Color-coded legends are provided for single animal identification.

• 

  FIGURE 7. Area under the receiver operating characteristic curves AC parameters were tested for ischemic detection. AUC, 95% confidence interval and p-values are included. (A) The significant single AC parameters (HR, PT, DT, S3DV) are here reported. (B) Combinatory models derived from logistic regression outperforming single parameters are reported.