R.W.F. Campbell

Recognition of ventricular fibrillation using neural networks

VENTRICULAR FIBRILLATION (VF) is a lethal cardiac arrhythmia which can be stopped by prompt application of a DC shock (defibrillation). On the electrocardiogram (ECG), VF appears as an irregular, undulating waveform. (Fig. 1) ECG monitoring systems and automatic defibrillators attempt to reduce delays in recognising VF by incorporating automatic detection algorithms. However, electrode artifact and other cardiac arrhythmias can produce VF-like ECG signals which lead to false positive detections. In addition, the ECG waveform during VF is poorly characterised and this can lead to false negative detections. Development and testing of VF detection algorithms has relied on a limited database of VF recordings because VF is unpredictable, and hence rarely recorded in a form suitable for analysis. Although many VF detection techniques have been developed and claim good performance (JAKOBSSEN e t al., 1990), independent evaluation has shown that some techniques are not optimal (CLAYTON et al., 1993). There is clearly room for further development. One approach with potential lies in the area of neural computing. Neural computing is a rapidly growing field (MILLER et al., 1992; BEALE and JACKSON, 1990). Simple processing units (neurons) are linked together by weighted connections. Each neuron processes its weighted inputs according to its activation function, and the output is then passed on to the inputs of the next layer of neurons. The remarkable feature of a neural network lies in its flexibility. By allocating appropriate values to the weights, a network can perform specific and complicated operations on its inputs. A network can hence be trained to perform a particular operation, using a set of training data comprising a series of input patterns for which the correct output is known. Each training pattern is repeatedly presented to the inputs. The network weights, orginally set to random values, are progressively optimised using a training algorithm which attempts to produce the correct output for the training patterns. Training continues until the errors associated with

Comparison of four techniques for recognition of ventricular fibrillation from the surface ECG

Four ventricular fibrillation (VF) detection techniques were assessed using recordings of VF to evaluate sensitivity and VF-like recordings to evaluate specificity. The recordings were obtained from Coronary Care Unit patients. The techniques were: threshold crossing intervals (TCl); peaks in the autocorrelation function (ACF); signal content outside the mean frequency (VF-filter); and signal spectrum shape (spectrum). Using 70 extracts, each 4 s long, from VF recordings, the VF filter achieved a sensitivity of 77 per cent; the ACF, TCl and spectrum algorithms had sensitivities of 67, 53 and 46 per cent, respectively. Susceptibility to false alarms was assessed using 40 extracts from VF-like recordings. The TCl algorithm was the most specific (93 per cent), while the spectrum, VF filter and ACF algorithms had specificities of 72, 55 and 38 per cent, respectively. The TCl algorithm achieved overall sensitivity of 93 per cent and specificity of 60 per cent. The spectrum, VF filter and ACF algorithms had overall sensitivities of 80, 93 and 87 per cent, and overall specificities of 60, 20 and 0 per cent, respectively.

COMPARISON OF PROCAINAMIDE AND MEXILETINE IN PREVENTION OF VENTRICULAR ARRHYTHMIAS AFTER ACUTE MYOCARDIAL INFARCTION

The incidence of ventricular arrhythmias after myocardial infarction has been compared in a controlled study of procainamide, mexiletine, and placebo. Sixty male patients who has sustained a myocardial infarction and had received lignocaine for ventricular tachycardia or ventricular ectopic beats which were R-on-T, multiform, or close-coupled took part. The efficacy of the drugs was evaluated by continuous 24-hour recordings of the electrocardiogram on the 4th and 10th days after admission to the study. Procainamide was given as 500 mg. 4-hourly and mexiletine as 250 mg. 8-hourly with corresponding placebo regimens for 12 days. 77% of patients receiving placebo showed serious ventricular rhythm disorders compared with 33% receiving antiarrhythmic therapy (p smaller than 0.05). Although only 35% of patients receiving procainamide achieved accepted therapeutic plasma concentrations compared with 95% of those receiving mexiletine, both drugs were equally effective antiarrhythmically. The only major adverse effect of therapy noted was development of a positive antinuclear factor in a procainamide-treated patient. These results demonstrate the efficacy of oral antiarrhythmic agents in the management of ventricular arrhythmias after acute myocardial infarction. Mexiletine has the advantage of less frequent administration and lower toxicity.

Mexiletine in the prophylaxis of ventricular arrhythmias during acute myocardial infarction

In a double-blind study involving 165 patients we examined the role of mexiletine, a new antiarrhythmic drug, for the prophylaxis of ventricular arrhythmias after acute myocardial infarction. Mexiletine or placebo was given orally to patients on arrival in the coronary care unit, and continuous electrocardiographic tape recordings were used to document arrhythmias. Ventricular tachycardia and R-on-T ventricular ectopic beats were significantly reduced in the mexiletine patients, but too few episodes of ventricular fibrillation occurred for statistical comment. When arrhythmias did occur in the mexiletine group, it was usually early in the study, at which time plasma drug levels were low. Adverse effects were uncommon. Patients who were given therapy, but in whom acute myocardial infarction could not be confirmed, suffered no serious consequences of taking mexiletine. The results demonstrate the benefit and limitations of prophylactic oral antiarrhythmic therapy for patients with acute myocardial infarction.

ESCWGA/NASPE/P EXPERTS CONSENSUS STATEMENT : LIVING ANATOMY OF THE ATRIOVENTRICULAR JUNCTIONS. A GUIDE TO ELECTROPHYSIOLOGIC MAPPING

Anatomic Nomenclature of the AV Junctions. Current nomenclature for the AV junctions derives from a surgically distorted view, placing; the valvar rings and the triangle of Koch in a single plane with anteroposterior and right‐left lateral coordinates. Within this convention, the aorta is considered to occupy an anterior position, whereas the mouth of the coronary sinus is shown as being posterior. Although this nomenclature has served its purpose for the description and treatment of arrhythmias dependent on accessory pathways and AV nodal reentry, it is less than satisfactory for the description of atrial and ventricular mapping. To correct these deficiencies, a consensus document has been prepared by experts from the Working Group of Arrhythmias of the European Society of Cardiology and from the North American Society of Pacing and Electrophysiology. It proposes a new, anatomically sound, nomenclature that will be applicable to all chambers of the heart. In this report, we discuss its value for description of the A V junctions and establish the principles of this new nomenclature.

Surgery for ventricular tachycardia associated with right ventricular dysplasia: disarticulation of right ventricle in 9 of 10 cases.

Ten patients (nine men, one woman; mean age 39 years) with arrhythmogenic right ventricular dysplasia underwent surgery to control life-threatening drug refractory ventricular arrhythmias. All had ventricular tachycardia causing syncope and six had a history of cardiac arrest. In all a minimum of three antiarrhythmic drugs (mean five) had been ineffective. At operation, the right ventricle was grossly diseased in all patients. Ventricular tachycardias were induced and mapped intraoperatively in all patients. The surgical plan was to ablate the arrhythmogenic focus if it was less than 4 cm2; one patient was so managed. Of the remaining nine, four underwent partial (approximately 40% of the right ventricular free wall) and five underwent total right ventricular disarticulation. All survived the operation and are alive at a mean follow-up interval of 24 months (range 5 to 67). Two patients developed new sustained ventricular tachycardias. These were well tolerated and, unlike the original arrhythmias, were easily controlled by drug treatment. All patients who underwent right ventricular disarticulation manifested signs of right heart failure in the early postoperative period, but these lessened progressively with the development of systolic septal movement into the right ventricular cavity. All 10 patients are in New York Heart Association class I or II at last review. In selected patients with arrhythmogenic right ventricular dysplasia, surgery offers a curative treatment for ventricular tachycardia and should be considered for patients whose arrhythmias are life-threatening and refractory to drug treatment.

Surgery for control of recurrent life-threatening ventricular tachyarrhythmias within 2 months of myocardial infarction☆

Twenty-seven patients (mean age 57 +/- 7 years) underwent surgery for control of recurrent drug-refractory ventricular tachyarrhythmias (uniform ventricular tachycardia alone in 9 patients, ventricular tachycardia and ventricular fibrillation in 15 and ventricular fibrillation alone in 3) within 2 months of acute myocardial infarction. The mean number of major arrhythmic episodes per patient was 15 (range 2 to 200) and of drug failures 4 +/- 2. Left ventricular function was severely impaired in the majority (ejection fraction 29%; range 14% to 47%) and 18 patients (66%) had a left ventricular aneurysm. Endocardial resection guided by a combination of endocardial activation mapping during tachycardia and fragmentation mapping during sinus rhythm was performed in all patients. All electrically abnormal left ventricular endocardium was excised. Eight patients (29.6%) died within 30 days of surgery. Death was not related to age, time of surgery after infarction, ventricular function, bypass time or type of arrhythmia. Patients requiring emergency surgery had a higher early postoperative mortality rate than did those undergoing planned surgery (43% versus 15%). During a follow-up period of 32 +/- 20 months, there have been no arrhythmic deaths and only three patients (16%) have required antiarrhythmic drug therapy. When required in the early weeks after infarction, surgery for ventricular arrhythmias offers a high cure rate at a risk related to the patients preoperative arrhythmia frequency, which in turn relates to the risk of arrhythmic death.

EFFECT OF A HIGHLY PURIFIED HYALURONIDASE PREPARATION (GL ENZYME) ON ELECTROCARDIOGRAPHIC CHANGES IN ACUTE MYOCARDIAL INFARCTION

A highly purified preparation of hyaluronidase (GL enzyme) was given in a double-blind, placebo-controlled, randomised study to 192 consecutive patients within 12 h of suspected myocardial infarction. Compared with those receiving placebo, patients with definite myocardial infarction given GL enzyme had significantly less change in QRS complexes; in those with anterior infarction the development of Q waves was less prominent. At 4 months the overall mortality among those with definite and possible myocardial infarction receiving GL enzyme (6 out of 83 patients, 7.2%) was lower than that in those receiving placebo (11 out of 79, 14%); this difference was not significant. No adverse effects were observed.

Evidence for Electrical Organization During Ventricular Fibrillation in the Human Heart

Electrical Organization During VF. Introduction: Ventricular fibrillation is a most dangerous cardiac arrhythmia that has received considerable attention, yet its pattern of electrical activation remains controversial. The aim of this study was to investigate the degree of organization during the clinical arrhythmia and to examine the phase relationship between deflections in independent ECG leads.

Accuracy of automatic QT measurement techniques

The aim of this study was to compare, against manual measurement, 3 algorithms for automatic QT interval measurement using eight 4 lead ECGs recorded from normal subjects. The algorithms investigated determined T wave end from (a) the intercept of the isoelectric line and maximum T wave slope, thresholds applied to (b) the T wave and (c) the differential of the T wave. Mean manual-automatic QT measurement differences of 27 to 29 ms for algorithm a, -2 to 20 ms for b and -5 to 4 ms for c were found. The standard deviations (SD) of the manual-automatic differences varied from 9 to 30 ms. Failure to measure the QT interval (1 to 12 cases) occurred with only algorithm b. Each algorithm performed differently with the mean and SD of the differences, and failure rate for b, dependent on threshold and isoelectric levels used.<<ETX>>