Electrocardiographic criteria for differentiating aberrancy and ventricular extrasystole in chronic atrial fibrillation: Validation by intracardiac recordings

J. ELECTROCARDIOLOGY 18 (1), 1985, 41-50 Electrocardiographic Criteria for Differentiating Aberrancy and Ventricular Extrasystole in Chronic Atrial Fibrillation: Validation by Intracardiac Recordings BY SAJAD GULAMIIUSEIN, M.D., F.R.C.P.(C), F.A.C.C.,* RAYMOND YEE, M.D.,I" PATRICK T. Ko, M.D., F.R.C.P.(C), F.A.C.C.** AND GEORGE J. KI.EIN, M.D., F.R.C.P.(C), F.A.C.C.~I" SUMMARY Many electrocardiographic criteria have been proposed for the differentiation of ven- tricular extrasystole and supraventricular conduction with aberrancy in atrial fibrilla- tion but the validity of these have not been confirmed by intracardiac studies. We record- ed His bundle electrograms in nineteen patients (eleven men, eight women) referred for diagnosis of abnormal QRS complexes in the context of chronic atrial fibrillation. Of 1,068 wide QRS complexes analyzed, 91% proved to be of ventricular origin. Electrocar- diographic criteria which were specific for ventricular extrasystole included: 1) left bun- dle branch block morphology, 2) right bundle branch block morphology with a monophasic R in lead V1 or an RS or QS pattern in lead V6, 3) presence of a "compen- satory pause", i.e., compensatory cycle (V2-V3) longer than the average cycle length of ten normally conducted beats preceding the abnormal complex (927 _ 317 vs 780 _ 199, mean _ SD in msec. p < 0.005), 4) frontal QRS axis of the abnormal complex directed superiorly or to the right and 5) the presence of a "short-long" cycle sequence. Right bun- dle branch morphology with a triphas.ic R in lead V1 or QRS pattern in V6 and concordant initial vector in lead V1 or in more than one ECG leads were very specific for supraven- tricular conduction with aberrancy. Analysis of coupling interval and Ashman's phenomenon, i.e., the long-short cycle sequence, were not specific for supraventrlcular conduction with aberrancy. We conclude that in digitalis-treated patients with chronic atrial fibrillation the ma- jority of abnormal QRS complexes are of ventricular origin. The diagnosis of ventricular extrasystole or aberrancy can be made using a single ECG lead (V1) and applying a com- bination of easily applied criteria. Atrial fibrillation is a commonly encountered clinical arrhythmia characterized by rapid, chaotic atrial depolarizations and an irregular ventricular response. Wide QRS complexes are From the Clinical Electrophysiology Laboratory, Depart- ment of Medicine, University Hospital and the University of Western Ontario, London, Ontario, and Section of Car- diology, Department of Medicine, University Hospital, Saskatoon, Sask. *Assistant Professor, Saskatoon, Saskatchewan. 1"Research Fellow, London, Ontario. **Assistant Professor, St. Johns, Newfoundland. j'tAssistant Professor, London, Ontario. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. w 1734 solely to indicate this fact. Reprint requests to: Sajad Gulamhusein, M.D., Section of Cardiology, Department of Medicine, University Hospital, Saskatoon, Saskatchewan, S7N OXO. frequently observed during atrial fibrillation and these may be supraventricular with aberrant in- traventricular conduction or ventricular ex- trasystole. This distinction may be more difficult in atrial fibrillation than sinus rhythm, since the diagnostic value of observing discrete atrial ac- tivity is lost. A number of surface electrocar- diographic criteria have been proposed to provide guidelines for the differentiation of wide QRS complexes during atrial fibrillation. 1"3 Although these criteria have been widely accepted by many clinicians, their usefulness and specificity remain to be established. We used intracardiac His bun- dle recordings in patients with chronic atrial fibrillation and frequent abnormal QRS com- plexes to evaluate the current electrocardiograhic criteria and establish newer criteria for differen- tiating supraventricular conduction with aber- rancy from ventricular extrasystole. 41 42 GULAMHUSEIN ET AL 18-43-21 I I I VO V1 V2 V3 3 ~ ,%_ ,- , - .__ ,'% 4 .~ v. "- 'V--'~-- ~ v~ .__.k._ L----,L.--- L - - "L . - - - HBEp - ~- - - - - - HBED "- ' -*~' '~" - - . 'e"- ' - - ": -~ I 500msec l i I I Fig. 1. Intracardiac His bundle elec- trograms recorded simultaneously with surface ECG leads during chronic atrial fibrillation. An abnormal QI{S complex, V2, is diagnosed as supraventricular with right bundle branch block mop phology as it is preceded by a Itis bundle electrogram (H) with an ItV interval identical to that of normally conducted beats, V0, V1 and V3. V0-V1 is the con- trol cycle length; V1-V2 the coupling in- terval and V2-V3 the compensatory cy- cle. The compensatory cycle following the supraventricular beat with abberran- cy is not prolonged. 1, 2, 3, V1, V6 = sur- face ECG leads HBED - - bipolar distal llis bundle electrogram HBEp - - bipolar proximal His bundle electrogram MATERIALS AND METHODS The patient population consisted of eleven men and eight women. After written informed consent was ob- tained, patients were brought to the electrophysiology laboratory and studied in the nonsedated, post- absorptive state. A single tripolar (#6 French) electrode catheter was introduced via the femoral vein and posi- tioned across the septal leaflet of the tricuspid valve to record a stable His bundle electgrogram. 4 This was recorded simulataneously with leads I, II, I I I , V1 and V6 of the surface ECG on an Elema ink-jet mingograph recorder at a paper speed of I00 mm/sec. Each patient had a recording period of 20 minutes. Def in i t ions and Data Ana lys i s : An abnormal QRS complex was diagnosed as supraventricular with aberrancy if the complex was preceded by a His bundle deflection with an H-V inter- val equal to or greater than that seen for complexes with normal QRS configuration (Fig. 1). The beat was otherwise considered ventricular in origin (Fig. 2). The following parameters were assessed for each ab- normal complex: a) QRS Morphology: The morphology of the abnor- mal QRS complex from the surface electrocardiogram 5 was categorized (Figs. 3, 4, 5) and the sensitivity and specificity of each morphologic type in differentiating aberrancy from ventricular extrasystole was assessed. b) Frontal Axis: Frontal axis of the abnormal QRS complex was calculated from the three limb leads using the hexaxial reference system and designated to one of the four quadrants (Fig. 6): south east quadrant -- nor- mal axis, defined as a frontal QRS axis in the range of -30 to +90; north east quadrant -- left axis deviation, an axis of -30 up to -90; north west quadrant -- superior axis, an axis in the range of -90 to _+ 180 and I z ---A,--- 5 Vl V' - - - v6 - - / i l ,V HBEo-- - - - , r 19-79-45 [ t -v . - , - - ,..-,.__._ ,,.r-'--- v"--" /, -4 J[,, v ' - - - - - A _ ;,v ", ill ",,.t" 500msec Fig. 2. Intracardiac His bundle recor- dings of ventricular extrasystole (arrow) with right bundle branch block mor- phology. Note that the ventricular ex- trasystole is not preceded by a ttis bun- dle electrogram (H). Also note the presence of a long compensatory cycle following the ventricular extrasystole. J. ELECTROCARDIOLOGY 18 (1), 1985 CHRONIC ATRIAL FIBRILLATION 43 LBBB PATTERNS OF QRS IN LEAD V 6 TYPE COMPLEX ABERRANT VENTRICULAR ORIGIN Q 2 329 Fig. 3. Patterns of left bundle branch block morphologies of the abnormal QRS complex in lead V6 during chronic atrial fibrillation. | | Q | 3 231 I 13 18 6 594 south west quandrant -- right axis deviation, an axis in the range of +90 to _ 180. The frequency with which aberrantly conducted supraventricular beats and ven- tricular extrasystoles were distributed in each quadrant was assessed. c ) In i t ia l QRS Vector: The initial vector {0.04 seconds) of the abnormal QRS complex was compared to the initial vector of a normally conducted beat in lead V1 and in all five ECG leads recorded. The diagnostic value of concordant initial vector was analyzed. d) Cycle lengths of beats preceding and following an abnormal QRS complex: The two normal beats im- mediately preceding the abnormal QRS complex, V2, were designated V0 and V1 and the normal beat suc- ceeding V2 was designated V3 {Fig. 1). The heats were used to define the following intervals and cycles: V0-V1 -- Control cycle V1-V2 -- Coupling interval V2-V3 -- Compensatory cycle V0-V1/V1-V2 -- ratio of the control cycle length to the coupling interval {"Ashman's Index") The coupling interval (V l-V2) was considered as fixed if the dr, ration varied I)y less than 120 reset. 6 The (liagnnstic usefulness of this criterion w~,s ~,,mlyze(I. The compensatory cycle (V2-V3) was compared to the average cycle length (ARR) of ten normally con- ducted beats immediately preceding each abnormal beat (V2) to assess the diagnostic value of the "com- pansatory pause". 7 Ashman's index was defined as the ratio of the COla- trol cycle to the coupling interval and arbitrarily divid- ed into four groups, equal to or greater than (>1 1.75. > 1.5 > 1.25 and > 1.00. The sensitivity and specificity of each ratio was assessed for differentiating aberrancy from ventricular extrasystole. Statistical Analysis The percentage differences between the two groups were analyzed using the Chi-square test. The statistical significance of matching paired data was compared using student's test for paired data. Standard definitions were used to calculate sensitivi- ty, specificity and predictive accuracy. Sensitivity (%): True positive X 100 True positive + false negative Specificity (%): True negative X 100 True negative + false positive Predictive Accuracy (%): True positive X 100 True positive + false positive RESULTS A total of 1068 abnormal QRS complexes were observed. Nineteen patients, mean age 56 years, range 22-78 years (Table I) with chronic atrial fibrillation were studied. Thirteen of the nineteen patients were on digoxin therapy {Table I). Nine hundred seventy-three (91.1%) of these were ven- tricular and only 95 (8.9%) were supraventr icular d. ELECTROCARDIOLOGY 18 (1), 1985 44 GULAMHUSEIN ET AL RBBB PATTERNS OF QRS IN LEAD V~ VENTRICULAR TYPE COMPLEX ABERRANT ORIGIN C) _ /~ ,s 57 | ~ 2 s | _~ 5s - (~) ~ I0 177 | . _~ - ,7 | _./L_ - ,2 82 S50 RBBB PATTERNS OF QRS IN LEAD V 6 VENTRICULAR TYPE COMPLEX ABERRANT ORIGIN @ ~ 27 - | _~ 5z ,,4 | ~ , ,89 | _ ~ - a | -@ - , 82 550 Fig. 4. Patterns of right bundle branch block morphologies of the abnormal QRS complex in lead V1 during chronic atrial fibrillation. Note that the triphasic R is found exclusively in aberrantly con- ducted supraventricular beats and the monophasic R is found predominantly with ventricular extrasystole. Fig. 5. Patterns of right bundle branch block morphology of the abnormal QRS complex in lead V6 in patients with chronic atrial fibrillation. Note that the QS or rs type of morphology is seen predominantly with ventricular ex- trasystole and qRs pattern with aber- rancy. J. ELECTROCARDIOLOGY 18 (1), 1985 CHRONIC ATRIAL FIBRILLATION 45 Fig. 6. Analysis of frontal QRS axis of abnormal complexes in patients with chronic atrial fibrillation. Note that most of the abberrantly conducted supraventricular beats, have a normal frontal axis. Superior QRS axis is seen only with ventricular extrasystoles with right bundle branch block morphology and right axis deviation is also more common in ventricular extrasystoles with right bundle branch morphology. Analysis of Frontal Axis SVBA RBBB -90 + 1 8 0 ~ ~ 3~ +90 LBBB 2/ PVB RBBB LBBB 5/ with aberrant intraventricular conduction (Table II). QRS Morphology: Ventricular extrasystole: The QRS morphology of 973 beats of ventricular origin (Table II) was analyzed. Sixty-three per- cent had left bundle branch block morphology (specificity 94%) and 37% had right bundle branch block morphology (specificity 6%) (Table II). The majority of complexes with left bundle branch block morphology had rSR or RSr pattern in lead V6 (Fig. 3). Of the ventricular beats with right bundle branch block morphology in lead V 1, Fig. 7. Schematic representation of the cycle, a) Long-short cycle sequence with V0-V1 greater than V1-V2 (Ashman's phenomenon); this was sensitive but not specific for aberrancy, b) Short-long cy- cle sequence with V0-V1 less than V1-V2; this was specific for ventrieular extrasystole but not sensitive. _Cycle Long-Short Cycle (Ashman's v0 Sequence Criteria Phenomenon) (V0 -V 1 >V 1 -V2) vl v2 b) Short-Long Cycle (V0-V 1 46 GULAMHUSEIN ET AL TABLE I Clinical Data Pt Age Sex Clinical Diagnosis Medications 1 22 M CAF 2 73 M CAF 3 66 M CAF 4 66 M CAF 5 67 F CAF 6 49 M CAF 7 78 F CAF 8 75 F CAF 9 31 M CAF 10 42 F CAF 11 62 F CAF, 12 62 M CAF, 13 73 F CAF, 14 56 M CAF, 15 65 M CAF, 16 41 F CAF, 17 72 M CAF 18 43 M CAF 19 70 F CAF, Friedrichs Ataxia Digoxin CAD Digoxin CAD Digoxin, Isordil COLD Digoxin, Diuretics MS Digoxin, Diuretics MI, AI Digoxin, Diuretics, Inderal SSS Hypertension Digoxin, Diuretics, Methyldopa SSS Hypertension Digoxin MS Digoxin no meds MVD no reeds CAD Digoxin MS no meds CAD no meds CAD Digoxin MS Digoxin no meds no meds MS, MI Digoxin, Quinidine Abbreviations: AI -- aortic insufficiency; CAD -- coronary artery disease; CAF -- chronic atrial fibrillation; COLD -- chronic obstructive lung disease; MI -- mitral insufficiency; MS -- mitral stenosis; SSS - sick sinus syndrome. 84 had a monophasic R, (sensitivity 24%, specificity 98%). 177 had a biphasic R with a taller left peak (sensitivity 51%, specificity 88%) and 57 had a biphasic R with a taller right peak (sensitivity 16%, specificity 84%) (Fig. 4). Right bundle branch block with QS and rS morphology in lead V6 (Fig. 5) appeared to be specific for ven- tricular extrasystole. Any conclusion on the specificity of other patterns is not possible because of the small numbers of aberrant supraventricular beats. Supraventricular beats with aberrancy: of the 95 aberrantly conducted supraventricular beats, 89 had right bundle branch block morphology (sensitivity 94%, specificity 37%) (Table II). Triphasic R in lead V1 (Fig. 4) and qRs pattern in TABLE II Distribution of Total Abnormal QRS Complexes 1068 ABNORMAL COMPLEXES 973 (91.1 %) 95 (8.9%) PVCs SVBA 617(63%) 356(37%) 89(94%) 6(6%) LBBB RBBB RBBB LBBB Abbreviations: PVCs -- premature ventricular contractions; SVBA -- supraventricular beat with aberrancy; LBBB -- Left bundle branch block. RBBB -- right bundle branch block. lead V6 (Fig. 5) were specific for aberrancy. Frontal Axis: Frontal QRS axis of 90 supraven- tricular beats with aberrancy (75 with right bun- dle branch block morphology and 15 with left bundle branch block morphology) and 971 ven- tricular extrasystoles (356 with right bundle branch block and 615 with left bundle branch block morphology) were analyzed. Supraventricular beats with aberrancy: The frontal QRS axis of nearly all the aberrantly con- ducted supraventricular beats were distributed to the south east (normal axis} or the north east (left axis} quadrants (Fig. 5). Only 1% of tlle supraven- tricular beats with right bundle branch block aberrancy were observed in the south west (right axis) quadrant. None were observed in the north west (superior axis) quadrant. Ventricular extrasystole: The distribution of frontal axis of ventricular extrasystoles with right bundle branch block morphology is shown in Figure 6. A frontal axis in tile north west quadrant (superior) or south west quadrant (right axis) was highly specific for ventricular ex- trasystole. Ventricular extrasystole with left bun- dle branch block morphology had predominantly normal (72%) or left axis deviation 27% (Fig. 6). Initial Vector: The initial vector of 806 abnormal complexes was analyzed. Seventy-two vectors were supraventricular with aberrancy and 734 were of ventricular origin. The overall sensitivity with which a concordant initial vector in a single J. ELECTROCARDIOLOGY 18(1), 1985 CHRONIC ATRIAL FIBRILLATION 47 precordial lead V1 predicted aberrancy was 89% with a predictive accuracy of 65%. The predictive accuracy was enhanced when the initial vector in all five surface ECG leads was concordant with the normally conducted beat (predictive accuracy 94%, sensitivity 68% -- (Table III). When we anaiyzed the initial vector of abnormal QRS com- plexes with right bundle branch block mor- phology alone, 96% of the aberrantly conducted supraventricular beats had a concordant initial vector in lead V1 and 78% had a concordant vec- tor in all five ECG leads (predictive accuracy of 91% and 100%, respectively). (Table III). Compensatory Pause: Ventricular extrasystole: The compensatory cycle (V2-V3) following a ven- tricular extrasystole was significantly longer than the average cycle length (ARR) of the ten normally conducted beats immediately preceding the ventricular extrasystole (927 + 317 vs 780 + 199 mean _+ SD in msec., P < 0.005). Supraventricular beats with aberrancy: There was no difference between the compensatory cy- cle length following aberrantly conducted supraventricular beats and the ARR (619 _+ 85 vs 631 _+ 98 mean __+ SD in msec., p NS). The positive predictive value of this criteria to differentiate ventricular extrasystole from aber- rancy was 91%. Coupling Interval: The coupling interval (V1-V2) was analyzed in 959 abnormal QRS complexes. Of these, 878 were of ventricular origin. Seventy- seven percent of ventricular and 81% of aberrant- ly conducted supraventricular beats followed a pattern of "fixed coupling". 6 The specificity of this criterion in differentiating ventricular ex- trasystole from supraventricular aberrancy was only 20%. Cycle Sequence Criteria: a) Effect of preceding cy- cle length (V0-V1): The length of the preceding cy- cle length V0-V1 did not aid in differentiating supraventricular beat with aberrancy from ven- tricular extrasystole. A long-short cycle se- quence, Ashman's phenomenon, (V0-V1 > V1-V2, Fig. 7a) was observed in 82% of aberrantly con- ducted "supraventricular beats. However the specificity of this criterion of only 7%. A short- long cycle sequence (V0-V1 < V1-V2, Fig. 7b) was more specific for ventricular extrasystole (96%), but the sensitivity of this criterion was only 4% (Table IV). b) Ashman's Index: The cycle-sequence criteria were also analyzed using the ratio of the control cycle length to the coupling interval (V0-VI/V1- V2), or the "Ashman's Index". The sensitivity and specificity of a ratio equal to or greater than 1.75 for diagnosing supraventricular beat with aberrancy were 47% and 48%, respectively. The sensitivity of this criterion improved in a step- wise fashion as the ratio was reduced but the specificity decreased (Table IV). DISCUSSION In chronic atrial fibrillation, it is important to make a distinction between ventricular ex- trasystole and supraventricular conduction with aberrancy, as its management depends on a cor- rect diagnosis. We found that 91% of all abnor- mal QRS complexes analyzed were of ventricular origin. This may reflect the presence of coexistent underlying heart and/or conduction disease (Table I) or may be related to digitalis use. None of our patients taking digitalis had clinical evidence of toxicity and the serum levels of digox- in were within the normal therapeutic range. Although others have suggested that ventricular extrasystole is more common than supraven- tricular conduction with aberrancy in chronic atr ial f ibri l lat ion 1 the overwhelming preponderance of ventricular extrasystole in this setting has not been previously demonstrated. In our study population, we did not observe any episodes of repetitive aberration s . In fact, episodes of ventricular bigeminy 9 and ventricular couplets were frequently noted. QRS Morphology: The recognit ion of characteristic QRS patterns is said to help in dif- ferentiat ing ventricular extrasystole from supraventricular beat with aberrancy. 1-3 As the right bundle branch is more susceptible to con- duction abnormalities than the left, an aberrantly conducted supraventricular beat would usually be expected to have right bundle branch block morphology. Our findings were consistent with this view. Of the 95 supraventricular beats with aberrancy observed, 89 {94%} had right bundle branch block morphology. However, right bundle branch block morphology was not specific for supraventricular conduction with aberrancy as 37% of ventricular extrasystoles also had a right bundle branch block pattern. Conversely, a left bundle branch block morphology was very specific for ventricular extrasystole (94%) in our study group. Other morphologic criteria have been described to differentiate abnormal QRS complexes in atrial fibrillation.l,2, 5 A triphasic QRS configuration in lead V1 has been said to favour the diagnosis of supraventricular beats with aberrancy whereas a monophasic or biphasic QRS configuration in lead V1 was considered sug- gestive of ventricular extrasystole. An rS or QS d. ELECTROCARDIOLOGY 18 (1), 1985 48 GULAMHUSEIN ET AL TABLE III ANALYSIS OF INITIAL ORS VECTOR 1 ) ALL ABNORMAL ORS COMPLEXES SENS. PA a) CONCORDANCE IN LEAD V I 89% 65% b) CONCORDANCE IN 5 SURFACE ECG LEADS (1,2,3, V 1, V 6) 68% 94% 2) QRS COMPLEXES WITH RBBB a) CONCORDANCE IN LEAD V 1 96% 78% b) CONCORDANCE IN 5 SURFACE ECG LEADS 91% 100% Legends Sens -- sensitivity PA -- positive predictive accuracy pattern in lead V6 suggested the diagnosis of ven- tricular extrasystole. 2 Our findings support the suggestions of previous authors 13,5 that an ab- normal QRS complex with a right bundle branch block and a triphasic (QRS) configuration in lead V1 or QRS configuration in lead V6 is specific for supraventricular conduction with aberrancy, whereas an abnormal complex with a right bundle branch block and a monphasic R in lead V1 or an RS or a QS pattern in lead V6 favours the diagnosis of ventricular extrasystole. -Initial Vector: Since initial ventricular activation (septal activation via the left bundle system} is unchanged in isolated right bundle branch block 1~ one would expect the initial vector of an aberrantly conducted impulse with right bundle branch block to be similar to the normally con- ducted impulses. Thus, concordance of the initial vector has been suggested as one of the major criteria :,2 to differentiate aberrancy from ven- tricular extrasystole. We found that less than 10% of ventricular extrasystoles with right bun- dle branch block morphology had a concordant in- itial vector in lead V1, and none had a concordant initial vector when all five surface ECG leads were considered. In contrast, 94% of aberrantly conducted supraventricular beats had a concor- dant initial vector in lead V1. The specificity of this criterion was increased when all five surface ECG leads were analyzed. Frontal Axis: An aberrantly conducted supraven- tricular beat has been associated with a normal frontal QRS axis whereas a bizarre, superior QRS axis, that is, one within the north west quadrant, has been said to suggest ventricular extra- systole. 2 In our study, aberrantly conducted supraventricular beats of both right and left bun- dle branch block had a normal or left axis devia- tion whereas right axis deviation and superior QRS axis were very specific for ventricular ex- trasystole. Thus, an abnormal QRS complex with a right bundle branch morphology and a superior or a right axis deviation is very suggestive of a ventricular extrasystole. Compensatory Pause: Langendorf suggested that a prolonged pause after an abnormal complex during atrial fibrillation suggested that the com- plex was a ventricular premature depolarization and termed this sign the "compensatory pause" of atrial fibrillation 7. The basis for this observa- tion was thought to be concealed retrograde con- duction into the atrioventricular node by ven- tricular premature depolarizations, leaving the atrioventricular node temporarily refractory to descending atrial impulses. Pritchett 11 and co- workers verified the validity of this concept. They showed that induced premature ventricular beats during atrial fibrillation in the laboratory were followed by a significant compensatory pause when mean data were considered, but felt that this would not be helpful in individual patients. Our study in patients with chronic atrial fibrillation and spontaneous ventricular ex- trasystoles further confirms Langendorf 's hypothesis regarding the presence of a compen- satory pause. In addition, we found that this elec- trocardiographic feature can be a useful diagnostic tool. The compensatory cycle (V2-V3) can be easily compared to the average cycle lengths of ten normally conducted beats (ARR) immediately preceding the abnormal complex. If this exercise is repeated for several identical ab- normal QRS complexes in a given patient, the compensatory cycle (V2-V3) will usually be greater than the average cycle lengths (ARR) when the abnormal beats are of ventricular origin. Coupling Interval: It has been suggested that fixed coupling is characteristic of ventricular ex- trasystole, whereas aberrancy is associated with variable coupling. 1.2 Allowing a variation be- tween coupling intervals of up to 120 msec. 6, we found that the specificity of this criterion was poor in differentiating ventricular extrasystole from aberrancy. This may be due to 1) the lenien- cy of the criterion utilized, 2) the tendency for ventricular aberrancy to occur after similar diastolic intervals, mimicking fixed coupling 9, or 3) the relatively high incidence of ventricular ex- trasystole with variable coupling. 6 Cycle Sequence Phenomenon: As the recovery phase of a given cycle is influenced by the length of the preceding cycle 12, aberration of supraven- tricular beat tends to appear if an impulse occurs early in diastole of the preceding beat before com- J. ELECTROCARDIOLOGY 18 (1), 1985 CHRONIC ATRIAL FIBRILLATION 49 .ASHMAN'S INDEX (Ratio of V0-VI/V1 -V2) >/1.75 >/1.5 >/1.25 >/1.0 Pt # Sens Spec Sens Spec Sens Spec Sens Spec 1 - - 100 - - 66 . . . . 2 - - 41 - - 6 - - 3 - - 50 GULAMHUSEIN ET AL REFERENCES 1. MARRIOTT, H J L AND SANDLER, I A: Criteria, old and new, for differentiating between ectopic ven- tricular beats and aberrant ventricular conduction in the presence of atrial fibrillation. Prog. Car- diovasc. Dis. 9:18, 1966 2. SCIIAMROTI[, L: Phasic aberrant ventricular con- duction. In Schamroth, L. ed. The disorders of car- diac rhythm, Vol I, 2nd edition. Blackwell Scien- tific Publication, 218, 1980 3. SANDLEa, I A AND MAmllOTT, H J L: The differen- tial morphology of anomalous ventricular com- plexes of right bundle branch block type in lead V, Ventricular ectopy versus aberration. Circulation: 31:551, 1965 4. SCnERLAG, B J, LAU, S H, HELPONT, R H, BERKOWITZ, W D, STEIN, E AND DAMATO, N: A catheter technique for recording His bundle activi- ty in man. Circulation 39:13, 1969 5. WELU.ENS, H J J, BAll, F W H M, VANAGT, S J D M AND BRUGADA, P: Medical treatment of ventricular tachycardia: Considerations in the selection of pa- tients for surgical treatment. Am J Cardiol 49:186, 1982 6. SURAWlCZ, B AND MacDoNALD, M G: Ventricular ectopic beats with fixed and variable coupling: In- cidence, clinical significance and factors influenc- ing the coupling interval. Am J Cardiol 13:198, 1964 7. LAN(;ENDOar, R: Aberrant ventricular conduction. Am Heart J 41:700, 1951 8. GouAux, J L AND ASUMAN, R: Auricular fibrillation with aberration stimulating ventricular parox- ysmal tachycardia. Am Heart J 34:366, 1947 9. LAN(;ENI)ORF, R, PicK, A ANI) IvVIN'I'ERNITZ, ~|*- Mechanisms of intermittent ventricular bigeminy; Appearance of ventricular ectopic beats dependent upon the length of the ventricular cycle; The rule of bigeminy. Circulation 11:422, 1955 10. GRANT, R P: Clinical electrocardiography. McGraw-Hill Book Co. Inc., New York, 1957, 119 11. PRITCIIETT, E L C, SMmx, W, KI.EIN, G J, HAM- MIl.l., S C AND GAX.LAt;m:It, J J: The "compen- satory pause" of atrial fibrillation. Circulation 62:1021, 1980 12. LEWIS, T AND MASTER, A M: Observations upon conduction in the mammalian heart: A-V conduc- tion. Heart 12:209, 1925 13. ASIIMAN, R AND BYEIt, E: Aberration in the con- duction of premature ventricular impulses. J. Loui- siana State Univ. School of Med. 8:62, 1946 J. ELECTROCARDIOLOGY 18 (1), 1985