The Retrograde P-Wave Theory: Explaining ST SegmentDepression in Supraventricular Tachycardia byRetrograde AV Node ConductionSANTIAGO RIVERA, M.D.,* MARIA DE LA PAZ RICAPITO, M.D.,* DIEGO CONDE, M.D.,†MARIANO BADRA VERDU, M.D.,* JEAN FRANCOIS ROUX, M.D.,*and FELIX AYALA PAREDES, PH.D.*From the *Division of Cardiology, Centre Hospitalier Universitaire de Sherbrooke, Sherbrooke, Canada;and †Division of Cardiology, Instituto Cardiovascular de Buenos Aires, Buenos Aires, Argentina

Background: Pseudo ischemic ST segment changes during supraventricular tachycardia (SVT) are notyet fully understood. Our aim was to determine whether venticulo-atrial (VA) conduction during SVT maybe a possible mechanism for ST depression (STd) in SVT.

Methods: Patients undergoing SVT ablation (2010–2012) were analyzed (n = 72).Typical atrioven-tricular node reentrant tachycardia (AVNRT) and atrioventricular reentrant tachycardia (AVRT) wereincluded. Those with STd were compared to those without STd. VA interval length, tachycardia cyclelength (TCL), and retrograde P-wave activation during SVT were assessed. Retrograde P waves arrivingsimultaneously with the ST segment (PWST) during SVT were considered, whenever an atrial electrogram(measured from the high right atrium) was “on time” with the ST segment.

Results: Patients with STd during SVT presented longer VA intervals than those without STd (VA100 ± 37 ms vs VA 69 ± 22 ms; P = 0.006). No differences in TCL were observed (TCL 333 ± 35 ms vsTCL 360 ± 22 ms; P = 0.1). PWST was observed in 38.5% of patients with AVNRT and STd versus 0% inthose without STd. The TCL was similar in both groups (355 ± 25 ms vs 334 ± 18 ms; P = 0.1). In patientswith AVRT and STd, PWST was present in 81% of cases versus 0% in those without STd. The TCL wasalso similar (330 ± 29 ms vs 346 ± 17 ms; P = 0.1).

Conclusions: STd during SVT is observed at long VA intervals when the retrograde P wave matches theST segment, without dependence on the TCL. This suggests that STd is not necessarily rate dependentbut a result of a fusion between the ST segment and the P wave. (PACE 2014; 37:1100–1105)

supraventricular tachycardia, ST depression, myocardial ischemia

IntroductionPronounced ST segment repolarization

changes are commonly observed in supraventri-cular tachycardia (SVT) but the baselinemechanism for this phenomenon is not yetfully understood.1–6 Indeed such ST-segmentdepression sometimes persists after the reductionof heart rate. The phenomenon was describedas posttachycardia ST depression. Fastertachycardias have been associated with positivetroponin values7 and no coronary artery disease,while coronary artery disease has proven not tobe related with ST depression (STd) in SVT.8The mechanism involved in STd during SVTremains unknown and it could be possible that

Address for reprints: Diego Conde, M.D., Blanco Encalada.,1543, Buenos Aires 1428, Argentina. Fax: 541147877500;e-mail: drconde@hotmail.com

Received January 15, 2014; revised January 26, 2010; acceptedFebruary 3, 2014.

doi: 10.1111/pace.12394

the cause is multifactorial. Regular, narrow QRStachycardia are commonly classified in long RP-short PR intervals and short RP-long PR intervals,suggesting that P waves can be localized at anypoint of the QT interval.9–13

We hypothesized that longer ventriculo-atrial(VA) conduction during SVT may be involved asa possible mechanism for STd in SVT, creating afusion image between the ST segment and the ret-rograde P wave in the 12-leads electrocardiogram(ECG). Our aim was to compare VA conductionin patients with SVT and STd and those withoutSTd, to assess whether retrograde atrial activationcould be related with this phenomenon.

MethodsStudy Population

Seventy-two patients undergoing an invasiveelectrophysiological study for SVT ablation wereassessed. All patients gave written informedconsent according to our institution guide-lines. Atrioventricular node reentrant tachycardia(AVNRT) and orthodromic reentrant tachycardia

©2014 Wiley Periodicals, Inc.

1100 September 2014 PACE, Vol. 37

RETROGRADE P-WAVES SIMULATING ST SEGMENT DEPRESSION

by accessory pathways (ORT) with QRS lessthan 100 ms were included. Only slow-fastAVNRT and ORT were included in this study.Patients with preexisting bundle branch block,ventricular repolarization anomalies, associatedarrhythmias (atrial tachycardia, atrial flutter, andatrial fibrillation), and preexcitation syndromes aswell as coronary artery disease and any coronaryrisk factors were excluded.

Electropysiologic Study

Three quadripolar catheters (Wolven,Josephson curve; Bard Electrophysiology, Lowell,MA, USA) were inserted into the right and leftfemoral veins and advanced into the high rightatrium (HRA), His, and right ventricular apex. Adecapolar catheter was placed into the coronarysinus via the right femoral vein. The surfaceECG leads and bipolar intracardiac electrogramswere recorded in the LabSystem PRO v2.6a (BardElectrophysiology) with filter setting between30 MHz and 500 MHz. Measurements wereperformed by two nonindependent observers at50 mm/s, 100 mm/s, and 200 mm/s speed. Themechanism of the tachycardia was determinedby standard techniques, including entrainmentmaneuvers, atrial and ventricular extrastimulustesting, and overdrive. Each patient was analyzedfor the tachycardia cycle length (TCL), VA intervallength, presence of STd during SVT, and whetherthe retrograde P wave, measured at the HRA,arrives at the time of the ST segment. TheTCL was measured as V to V recorded by aquadripolar catheter placed at the right ventricularapex. The VA interval was measured from thebeginning of the QRS to the latest atrial activationrecorded at HRA catheter. P wave matches STsegment (PWST) was considered if the latest atrialelectrogram (HRA) was “on time” with the STsegment recorded in the 12-lead ECG. We namedthis association: “P wave matches ST segment”(PWST). ST segment depression was defined as�2-mm depression persisting 80 ms after the Jpoint in >2 consecutive leads. An ST segmentscore (STSS) was calculated for each patient,adding the amount of STd (mV) in each lead andcompared in both AVNRT and atrioventricularreentrant tachycardia (AVRT).

Statistical Analysis

A test of normality was performed if consid-ered necessary. All data are expressed as meanand STd. Statistical analysis was performed usingthe t-test for unpaired observations and categoricalvariables were compared using the χ 2 or Fisher’sexact test. A P value of 0.05 or lower wasconsidered statistically significant.

Table I.

Differences between STd and No STd during SVT

No STd STd P =

AGE 48 ± 19y/o 45 ± 17y/o 0.5FEMALES 54% 58% 0.7TCL 360 ± 22ms 333 ± 35ms 0.1VA INTERVAL 69 ± 22ms 100 ± 35ms 0.006

STd = ST depression; SVT = supraventricular tachycardia;TCL = tachycardia cycle length; VA = venticulo-atrial.

ResultsA total of 72 patients were analyzed; 44.4%

were males and 56.6% females. The mean age was48 ± 16 years old; 76.4% had AVNRT and 23.6%AVRT. We observed ST segment depression in33.3% cases. The mean TCL was 347 ± 41 msand the mean VA interval was 79 ± 27 ms. Asmall number of patients (8.3%) presented fastertachycardia with TCL less than 280 ms. Patientswith ST segment depression during SVT hadlonger VA intervals than those without ST segmentdepression (VA 100 ± 37 ms vs VA 69 ± 22 ms;P = 0.006) and no difference in TCL (TCL 333 ±35 ms vs TCL 360 ± 22 ms; P = 0.1), as shown inTable I.

Retrograde P-Wave Matching ST Analysis

Patients with AVNRT diagnosis and STdduring tachycardia presented retrograde PWST in38.5% cases (Figs. 1 and 2), while no match (0%)was observed in those without STd (Figs. 3 and4; Table II). The TCL was similar in both groups(355 ± 25 ms vs 334 ± 28 ms; P = 0.1). Patientswith AVRT diagnosis and STd during tachycardiapresented retrograde PWST in 81% cases, whileno match (0%) was observed in those withoutSTd. The TCL was also similar in both groups(330 ± 29 ms vs 346 ± 17 ms; P = 0.1).

STSS Analysis

No significant difference was observed in theST score when we compared AVNRT and AVRTpatients (0.9 ± 0.2 mV vs 0.95 ± 0.15 mV; P = 0.5).A small group of patients presenting shorter TCL(<280 ms) presented a slightly higher ST score of1.2 ± 0.18 mV versus 0.9 ± 0.2 mV in those withlonger TCL (>280 ms).

DiscussionMechanisms Proposed for STd

Coronary artery occlusion has a low preva-lence and was not correlated to STd duringSVT,14,15 suggesting that marked STd during SVT

PACE, Vol. 37 September 2014 1101

RIVERA, ET AL.

Figure 1. 12 leads ECG showing an AVNRT with STd and TCL of 280 ms at 25 mm/s. AVNRT =atrioventricular node reentrant tachycardia; ECG = electrocardiogram; STd = ST depression;TCL = tachycardia cycle length.

Figure 2. Intracardiac activation sequences during slow-fast AVNRT at 100 mm/s. The VAinterval measured from the QRS onset to the proximal CS atrial electrogram is 23 ms. The VAinterval measured from the QRS onset to the latest atrial electrogram recorded at the HRA is 108ms. The latest retrograde atrial activation occurs after ventricular depolarization (QRS) and “on-time” with the ST segment during SVT. We referred to this as P wave matching the ST segment.AVNRT = atrioventricular node reentrant tachycardia; CS = coronary sinus; HRA = high rightatrium; SVT = supraventricular tachycardia.

1102 September 2014 PACE, Vol. 37

RETROGRADE P-WAVES SIMULATING ST SEGMENT DEPRESSION

Figure 3. 12 leads ECG showing an AVNRT with no STd and TCL of 295 ms at 25 mm/s.Abbreviations as in previous figures.

Figure 4. Intracardiac activation sequences during slow-fast AVNRT at 100 mm/s. The VAinterval measured from the QRS onset to the proximal CS atrial electrogram is 5 ms. The VAinterval measured from the QRS onset to the latest atrial electrogram recorded at the HRA (Abldist) is 47 ms. The latest retrograde atrial activation occurs during ventricular despolarization(QRS) without matching the ST segment in SVT. Abbreviations as in previous figures.

PACE, Vol. 37 September 2014 1103

RIVERA, ET AL.

Table II.

P-wave correlation with ST segment and TCL

No STd STd P=

AVNRT TCL 355 ± 25ms 334 ± 18ms 0.1PWST Match 0% 38% 0.001

AVRT TCL 346 ± 17ms 330 ± 29ms 0.7PWST Match 0% 81% 0.002

AVNRT = atrioventricular node reentrant tachycardia; AVRT = atrioventricular reentrant tachycardia; PWST = P wave matches STsegment; STd = ST depression; TCL = tachycardia cycle length.

may per se be the cause for STd by alteringthe slope of phase two of the ventricular actionpotential.16 The myocardium action potentialduration (APD) and effective refractory period(ERP) are longer at the epicardium at long cyclelengths, but a crossover of the relation is oftenencountered so that at short cycle lengths the APDand ERP become longer at the endocardium. In thepresence of very short TCL with higher oxygendemands, functional ischemia is present, andendocardial APD could be reduced beyond thatof the epicardium, thus generating dispersion ofrefractoriness accounting for STd in patients withno coronary artery disease (CAD)17–19 and veryfast tachycardia. This reveals that the STd may bemultifactorial.

Findings

Every patient included in this study hadno history of coronary artery disease and nocoronary risk factors. However, more than 30%developed STd during SVT. ORT had higher STdprevalence. No TCL differences were observedbetween AVRT and AVNRT, and only longerVA interval lengths and PWST were associatedwith STd. Short VA intervals will show P wavesmasked by the QRS while very long VA intervalswill show a P wave fused with or masked by the Twave, depending on the T-wave amplitude andlength. Very long VA intervals showed P-waveand T-wave fusion, giving an electrocardiographicpattern resembling a positive-negative T wave,with no STd. Intermediate VA intervals will eithershow a pseudo r/s pattern if the P wave is fusedwith the terminal aspect of the QRS or willinduce STd if the P wave is fused with the STsegment (Fig. 4). The “PWST” variable was usefulin determining the real association between the Pwave and the ST segment.

Mechanism Proposed for STd

This study suggests that retrograde VAconduction and PWST were determinants for STd.We believe that retrograde atrial activation can be

fused with the ST segment or T wave dependingon whether the retrograde P wave falls in the STsegment or later in the QT interval. This wouldexplain why tachycardia’s presenting longer VAintervals (ORT) have a higher prevalence of STdthan those with shorter or negative VA intervals(AVNRT).

Long VA Intervals in Typical AVNRT

High atrial activation can be longer thanproximal coronary sinus activation in patientspresenting STd during typical AVNRT. In allAVNRT cases the VA interval measured to theproximal CS always remained less than 70 ms.Intraatrial conduction delays during SVT mayexplain longer VA in AVNRT, accounting forlonger VA intervals and STd.

Study LimitationsAlthough VA length was the main determi-

nant of STd, we observed in a very small numberof patients with SVTs at very short TCL (lessthan 280 ms), that faster rates might give riseto STd independently of the VA interval length.This phenomenon would be secondary to func-tional ischemia by increasing myocardial oxygendemands, thus shortening the endocardial APDand developing transmural gradients. Patientsincluded in this study had no history of CAD. Eventhough coronary angiography or treadmill test wasnot systematically performed, this was a very low-risk population.

ConclusionsRetrograde P-wave activation arriving at the

same time of the ST segment was highly associatedwith STd. Long VA intervals in narrow QRS SVTaccount for STd when the retrograde P wavematches the ST segment. In every case, STdoccurred independently of the TCL, suggestingthat STd is not rate dependent. Tachycardia usingaccessory pathways presented STd significantlymore often than AVNRT as VA intervals tend tobe longer.

1104 September 2014 PACE, Vol. 37

RETROGRADE P-WAVES SIMULATING ST SEGMENT DEPRESSION

References1. Imrie JR, Yee R, Klein GJ, Sharma AD. Incidence and clinical signif-

icance of ST segment depression in supraventricular tachycardia.Can J Cardiol 1990; 6:323–326.

2. Riva SI, Della Bella P, Fassini G, Carbucicchio C, Tondo C. Value ofanalysis of ST segment changes during tachycardia in determiningtype of narrow QRS complex tachycardia. J Am Coll Cardiol 1996;27:1480–1485.

3. Slavich G, Pavoni D, Badano L, Popiel M. Significance of ST-segment depression during supraventricular tachycardia. Cluesoffered by its return to normal at the end of the episode. Ital Heart J2002; 3:206–210.

4. Takayanagi K, Hoshi H, Shimizu M, Inoue T, Sakai Y, MorookaS, Takabatake Y. Pronounced ST-segment depression duringparoxysmal supraventricular tachycardia. Jpn Heart J 1993; 34:269–278.

5. Akgun G, Diker E, Kaymakalan S, Kervancioglu C. Rate-dependentST segment depression–a case report. Angiology 1989; 40:844–848.

6. Nelson SD, Kou WH, Annesley T, de Buitleir M, MoradyF. Significance of ST segment depression during paroxysmalsupraventricular tachycardia. J Am Coll Cardiol 1988; 12:383–387.

7. Ben Yedder N, Roux JF, Paredes FA. Troponin elevation insupraventricular tachycardia: Primary dependence on heart rate.Can J Cardiol 2011; 27:105–109.

8. Gulec S, Ertas F, Karaoouz R, Guldal M, Alpman A, Oral D. Valueof ST-segment depression during paroxysmal supraventriculartachycardia in the diagnosis of coronary artery disease. Am J Cardiol1999; 83:458–460 A10.

9. Katritsis DG, Camm J. Atrioventricular nodal reentrant tachycardia.Circulation 2010; 122:831–840.

10. Nakagawa H, Jackman WM. Catheter ablation of paroxysmalsupraventricular tachycardia. Circulation 2007; 116:2465–2478.

11. Schamroth L. How to approach an arrhythmia. Circulation 1973;47:420–426.

12. Otomo K, Gonzalez MD, Beckman KJ, Nakagawa H, Becker AE,Shah N, Matsudaira K, et al. Reversing the direction of pacedventricular and atrial wavefronts reveals an oblique course inaccessory AV pathways and improves localization for catheterablation. Circulation 2001; 104:150–156.

13. Anderson RH, Ho SY, Gillette PC, Becker AE. Mahaim, Kent andabnormal auriculoventricular conduction. Cardiovasc Res 1996;31:480–491.

14. Dorenkamp M, Zabel M, Sticherling C. Role of coronary angiog-raphy before radiofrequency ablation in patients presenting withparoxysmal supraventricular tachycardia. J Cardiovasc PharmacolTher 2007; 12:137–144.

15. Petsas AA, Anastassiades LC, Antonopoulos AG. Exercise testing forassessment of the significance of ST segment depression observedduring episodes of paroxysmal supraventricular tachycardia. EurHeart J 1990; 11:974–979.

16. Litovsky SH, Antzelevitch C. Transient outward current prominentin canine ventricular epicardium but not endocardium. Circ Res1988; 62:116–126.

17. Lin YJ, Tai CT, Chiang CE, Lee KT, Yuniadi Y, Huang BH, LiuTY, et al. Mechanism of repolarization change during initiationof supraventricular tachycardia. J Cardiovasc Electrophysiol 2004;15:1233–1237.

18. Benito B, Cuasch E, Rivard L, Nattel S. Clinical and mechanisticissues in early repolarization. JACC 2010; 56:1177–1186.

19. Antzelevitch C, Sicouri S, Litovsky SH, Lukas A, KrishnanSH, Di Diego JM, Gintant GA, et al. Heterogeneity betweenthe ventricular wall. Electrophysiology and pharmacology ofepicardial, endocardial, and M cells. Circ Res 1991; 69:1427–1449.

PACE, Vol. 37 September 2014 1105