History of Hypertrophic Cardiomyopathy (HOCM), with Tachycardia and High Lactate

A patient under 40 with h/o HOCM and implantable cardioverter-defibrillator (for secondary prevention of VF arrest that occurred during exertion) presented with chest pain, diaphoresis, and tachycardia.  Earlier in the day, the patient had been physically active, which resulted in dizziness, SOB and diaphoresis.  Then later, there was alcohol consumption associated with further physical exertion.  The patient presented clutching the chest, dizzy, SOB, diaphoretic.  BP was 165/109 (a good example of shock in which the BP is maintained by high systemic vascular resistance).

Here is her ED ECG:

There is a narrow complex (QRS duration =113 ms) tachycardia at a rate of about 160.  There are no definite P-waves.  It appears to be paroxysmal SVT.  There is very high voltage and secondary repolarization (ST-T) abnormalities.

The Bedside Cardiac Ultrasound is shown here:


This shows the very hypertrophic walls, and the consequent very small left ventricular chamber collapsing on itself.  There is very little opporunity for the heart to fill with blood, and probable obstruction of aortic outflow as well.

The patient was given adenosine 6 mg, 12 mg, 12 mg, and 18 mg without any lasting effect.

Lactate returned at 8.3 mEq/L, consistent with shock.

Then an esmolol double bolus (each bolus = 500 mcg/kg) and drip was given, with immediate slowing of heart rate.  The patient was given a normal saline fluid bolus.

The heart rate immediately slowed and here is the ECG at t = 45 min:

Sinus tach with heart rate about 100.  High Voltage and secondary ST-T abnormalities.

The patient stabilized with a BP of 107/58 and pulse of 100 and felt much better.

Later that day:

Sinus rhythm.  LVH with typical repolarization abnormalities.  Though this ECG is diagnostic of LVH, it is not specifically diagnostic of HOCM.  The signatures of HOCM, besides LVH, are tall R-waves in septal leads (septal hypertrophy) and deep S-waves in lateral leads (reciprocal to septal hypertrophy, such as in the ECG from another patient below. 

Here is an example of HOCM with septal R-waves in the right precordium (different patient):

There is massive voltage with corresponding repolarization abnormalities.  There is very high voltage R-waves in right precordial leads, highly suggestive of septal hypertrophy.  The ST segments are depressed, discordant to the large R-wave.  This is unlike most LVH, in which the right precordial S-waves are deep, with discordant ST elevation.)  This is highly suspicious for hypertrophic cardiomyopathy with asymmetric septal hypertrophy (HOCM).   To see this entire case, with ultrasounds, go to this post.

Back to this case:

Interrogation of the IVCD revealed that the heart rate gradually slowed after esmolol (there was no sudden conversion of rhythm to suggest re-entry).  Thus, the initial ECG is believed to have had a rhythm of sinus tachycardia.

Whether sinus tach or SVT, beta blockade is an ideal therapy in this situation, along with fluids.  The LV's pump function is too vigorous, causing collapse.  And the fast heart rate leaves no time for cardiac filling.  The left atrial pressure is too low to allow good preload of the LV.  Therefore, one should increase preload by increasing left atrial pressure (by giving fluids), and both 1) slow the heart rate to allow for better filling and 2) decrease pump function to prevent obstruction of outflow and also allow for better filling (by giving beta blockade, in this case esmolol, as short-acting beta-1 blocker that can be discontinued if there are adverse events).

The high lactate shows how this patient is very volume dependent.  Dehydration can set off a spiral of low stroke volume, then further catecholamine output with consequent increased myocardial contractility, and thus LV chamber collapse, with subsequent obstruction of outflow and worse filling.  The vicious cycle needs to be broken.

Outcome:

The patient did very well and was instructed to not let herself get dehydrated.  It is uncertain if she was discharged on a beta blocker, but this is one potential therapy to help prevent recurrence.

A 29 year old male with pleuritic chest pain for 6 hours

This case was sent to me by Taylor Sanders of the LSU- Baton Rouge Emergency Medicine residency.

A 29 year old complained of 6 hours of pleuritic chest pain:

QRS: There is rSR' in V1, consistent with RV conduction delay, but the QRS does not appear prolonged and there are no S-waves in lateral leads.  However, this absence of lateral S-wave may be due to terminal QRS distortion from the ST Elevation.

R-waves in Lateral leads: the R-waves in I and aVL are minimal, but are well formed in V5 and V6; this is somewhat unusual, and one must entertain the possibility of reversed limb lead placement.  However, when the axis is 90 degrees, because V5 and V6 are inferior to limb leads I and aVL, they may show R-waves when I and aVL do not.   

PR segments: there appears to be some PR depression in leads II and V3.  In lead II, it is partly due to a downsloping baseline.  This PR depression is suggestive of myo-pericarditis, but may also be found in MI.  PR depression of greater than 0.8 mm is generally considered specific for pericarditis, but the data upon which this is established comes from the pre-angiogram era, and cannot be fully trusted.

ST segments: 

There is marked ST elevation in inferior and lateral leads.  Inferolateral STEMI and pericarditis are very difficult to distinguish, and the best means to do so is that with inferolateral STEMI there is virtually always ST depression in aVL, even when there is ST elevation in V5 and V6.  There is no reciprocal ST depression in aVL and this makes inferior STEMI, even with superimposed lateral STEMI, very unlikely. 

The ST axis in pericarditis is rarely to the right of lead II.  An ST axis towards lead II results in no limb lead ST depression except in aVR (which is the opposite of a lead midway between I and II)

The ST axis in inferior STEMI is almost always to the right of lead II (resulting in ST depression in aVL) 

ST depression in aVR is found in all etiologies of inferior ST elevation: pericarditis, early repol and MI.  So this is not helpful.  However, there is ST depression in V1 and V2: is this posterior injury?  Or are these ST segments simply discordant to the R' wave, and a result of the abnormal depolarization ("secondary", not "primary" ST segment abnormalities)?  In general, myo-pericarditis does not have reciprocal ST depression anywhere except lead aVR, but this may be an exception due to the R' waves in V1 and V2.

T-waves: These are prominent, and worrisome for MI or early repolarization.  Early repolarization is possible, but less likely when not also seen in anterior leads.  Slow upstroke and fast downstroke, and relatively short QT favor a non-MI diagnosis as well.  T-waves in II, III, aVF and V4-V6 are very tall relative to the R-wave and QRS.

QTc interval: I do not have the computerized QTc for this case, but my visual inspection puts it at no longer than 380 ms.  In our as yet unplublished study of inferior STEMI, only 7% had a QTc less than or equal to 380ms. 

I was shown this ECG with just the clinical information above.  I responded that this is probably myo- or peri-carditis but that I was worried about the size of the T-waves.   And I often say the "you diagnose pericarditis at your peril."  The fact that the pain was pleuritic and the patient was 29 years old is supportive of pericarditis, but look at this case of a 24 year old woman with chest pain after a night of drinking and vomiting.


The first troponin I returned at 6 ng/mL (significantly elevated).


Another ECG was recorded:

Now there is much more ST elevation in lateral precordial leads.  There is also STE in aVL.  (ST axis is to the left of lead II; this is highly suggestive of pericarditis.  But how much do you want to bet?)

These are dynamic ST segments.  Are ST segments in pericarditis this dynamic?  I don't think we have a lot of data on this.  But this is very worrisome for STEMI, so the patient was taken emergently to the cath lab.

A very reasonable course of action would be a formal echocardiogram to look for wall motion abnormalities.  Absence of WMA would make STEMI impossible and establish the diagnosis of pericarditis without subjecting the patient to an angiogram.

It is also very reasonable to avoid any delay and go directly to the cath lab.


All coronary arteries were clean.

Final diagnosis:  Myocarditis

Respiratory Failure and ST Depression: Is there Posterior STEMI?

The ultrasound in this case was recorded by Dr. Robert F. (Rob) Reardon, one of my partners here at Hennepin County Medical Center (HCMC) in Minneapolis, and one of the world leaders in emergency ultrasound.  He is also an editor of this great new textbook of emergency ultrasound (Ma, Mateer, Reardon, Joing, eds.), and one of the authors of the Cardiac Ultrasound chapter (other authors of this chapter are Dr. Andrew Laudenbach (also of HCMC) and Dr. Scott Joing (also of HCMC, and the creator of the outstanding FOAMed site, www.hqmeded.com).

Case

A middle-age woman with a history of emphysema presented in severe respiratory distress and respiratory failure.  She was intubated emergently in the ED.  Her venous blood gas after intubation had a pH of 7.16 and pCO2 of 66.  The Chest X-ray was suggestive of pneumonia, but not pulmonary edema.  The following ECG was recorded:

There is sinus tachycardia, and ST depression that is maximal in V3 and V4, suggestive of posterior STEMI, or possibly subendocardial ischemia.  [However, subendocardial ischemia is usually diffuse, and therefore has an ST depression vector towards the apex of the heart (towards V5 and V6.  That is to say, the maximal ST depression is usually in I, II, V5, and V6, with reciprocal ST elevation in aVR.]

A posterior ECG was recorded:

There is ST elevation in posterior leads V7 and V8.  

Although this meets criteria for posterior STEMI (0.5 mm in 2 leads), there will virtually always be some ST elevation in posterior leads when there is ST depression in anterior leads, as these are opposing leads. 

[There is an exception to this rule, and that would be in pericarditis, when there is an ST elevation vector that goes from endocardium to epicardium throughout the entire heart, with an ST elevation summation vector towards the apex.  In such a case, there is diffuse ST elevation, including towards the posterior wall.]

Thus, there is probably posterior transmural ischemia.  Is this ACS with posterior MI?  The presentation of respiratory failure without pulmonary edema is not at all typical for ACS.  The patient apparently has a COPD exacerbation with pneumonia.  She could have 2 pathologies at once, but this is less likely.

An ED cardiac echo was performed at the bedside:

This subcostal view shows poor contractility at the entire base of the heart, and excellent contractility at the apex.  There is no wall motion abnormality in a coronary distribution.

Dr. Reardon made a diagnosis.  What is it?












Reverse Takotsubo!  (See below for description of Takotsubo and Reverse Takotsubo)

Case continued:

The patient was admitted to the Medical ICU.  She recovered.  Her max troponin I was 2.2 ng/mL.  Formal Echo also showed Reverse Takotsubo, with EF of 35%.  Echo 2 months later showed full recovery of EF.

She returned in respiratory distress 5 months after the first presentation, and required intubation again.  Here is her ECG from that visit:

Very concerning for Anterior STEMI

A bedside echo showed what appeared to be an anterior wall motion abnormality.  Cardiology was immediately consulted for a formal echocardiogram.  It showed an EF of 15% with a circumferential loss of function at the mid-section, with preservation of the apex and the base.  (This is called mid-ventricular stress cardiomyopathy).

Again, the troponin I peaked at 2.2 ng/mL.

An angiogram was done and showed normal coronary arteries.

The LV function eventually recovered again.





Stress Cardiomyopathy, Takotsubo and Reverse Takotsubo, and Mid-Ventricular Takotsubo Cardiomyopathy

In Takostubo stress cardiomyopathy, caused by small vessel ischemia from high catecholamine influence, there is poor contractility at the apex, causing "apical ballooning," which has the appearance of a Japanese octopus trap, or "Takotsubo"  Here is a left ventriculogram of Takotsubo SCM.

Standard Takotsubo with Apical Ballooning.

See this case for ECG and Echo video of Takotsubo Stress Cardiomyopathy that Mimics STEMI.


Reverse and Mid-Ventricular Takotsubo Stress Cardiomyopathy (SCM):

Reverse Takotsubo SCM is the term used when the LV dysfunction is of the base, and not the of the apex.  Thus, there is no apical ballooning. As in standard Takotsubo, the dysfunction is circumferential, not in a vascular territory, and not due to ACS.  

Reverse SCM has been described in many stressful situations, just as standard Takotsubo SCM, including sympathomimetic drug abuse, energy drinks, serotonin syndrome, anaphylaxis, high dose epinephrine (adrenaline!), pheochromocytoma, subarachnoid hemorrhage, sepsis, and dobutamine stress.

Reverse Takotsubo may be more common in younger patients, but there is little systematic data on the condition.  One small registry of 103 SCM patients, 20 of whom had reverse Takotsubo, showed that the reverse type had higher incidence of triggering stress (100% vs. 77%), less dyspnea, pulmonary edema, and cardiogenic shock, and less T-wave inversion on ECG.

Mid-ventricular Takotsubo is the term for poor function of the mid LV (ballooning of the mid-LV), with good function of BOTH the base and the apex.  It is less common than either of the other forms.

Of course, if it is SCM that does not have apical ballooning, it does not look like an octopus trap, and therefore perhaps should not be called Takotsubo at all.

There is also a claim of a 4th type, "localized" SCM (with focal wall motion abnormalities mimicking ACS).  The claim is substantiated only by case reports, such as this one, which cannot establish with certainty the absence of a thrombotic coronary lesion.

Take Home Lesson:

When the clinical situation is stress (such as respiratory failure from COPD in this case -- not from pulmonary edema), and the echocardiogram shows circumferential dysfunction, whether at the base, mid-LV, or apex, then stress cardiomyopathy is very likely the etiology of the ECG abnormalities.

ED Bedside Echo for the last very subtle inferior MI case. Images obtained and interpreted by a novice.

This is from the last case of trapezius pain and syncope.  Here is the first ECG:

ST depression and T-wave inversion suggestive of inferior MI.  There is minimal ST elevatiion in inferior leads, and a T-wave in III that is large compared to a small QRS. But the ECG is nondiagnostic.

A pediatric MD who is learning bedside echo and had only performed 20 cardiac ultrasounds in his life was sent in to practice on this patient, not expecting to find anything, or even to be able to find anything.  He recorded this echo from the apical 4-chamber view:

Echo 1 done by newbie from Stephen Smith on Vimeo.


This 4 chamber view was recorded by an emergency echo expert:

Echo 13 from Stephen Smith on Vimeo.


Here is a still frame with a circle that shows the area of wall motion abnormality:

This is taken in systole.  The base of the septum (at the left of the circle) is not contracting as is the more apical portions of the septum.

Here is a two chamber view (LV and LA only, with the probe turned 90 degrees from the apical 4-chamber) which shows the infero-posterior wall on the left side of the image.

2 chamber view from Stephen Smith on Vimeo.

Here is a still frame in systole showing the wall motion abnormality at the base of the heart on the left side of the picture:

Here is a map of the echocardiographic wall motion abnormalities:

Reproduced with permission of Robert F. Reardon, from Ma, Mateer, Reardon, and Joing, Editors.  Emergency Ultrasound.  McGraw Hill 2014.  The chart shows the coronary supply of the segments of the heart as seen on various views.  The circle shows the area of wall motion abnormality in this patient's apical 4 chamber view.  The lavendar on the "long axis" 2 chamber view correlates with this patient's 2 chamber view.  Both are consistent with the RCA involvement, and thus correlates perfectly with the ECG.

This shows how even a novice may be a able to see wall motion abnormalities that, especially if they correlate with the subtle ECG findings, may help to make the diagnosis of ACS.

Warning:

I would warn that wall motion abnormalities may be very difficult to see, and that there are also false positives.

It frequently requires echo contrast, and an expert echocardiographer and interpreter to accurately obtain and read the images.

But there are times when those with less experience can provisionally identify wall motion abnormalities, and it may help in diagnosis of ACS.

Sudden Left Trapezius Pain and Syncope

An otherwise healthy middle aged smoker complained of sudden left trapezius pain.  He then began to fell very dizzy.  He also had some vague left sided chest pain.  His prehospital ECG was identical to the first ED ECG:

There is minimal ST elevation in inferior leads.  What caught my eye was the T-wave inversion and minimal ST depression in aVL.  This should always make you scrutinize the inferior leads for signs of inferior MI, and record serial ECGs.  However, to my eye, inferior ST-T complexes did not appear to show MI.

The patient's symptoms were rather atypical, and because of this and the benign appearing inferior leads, I thought that the ST depression and T-wave inversion in aVL might not be pathologic.  I signed the patient out to the next team with a plan to get serial ECGs and troponins, and manage according to the results.

The patient received NTG without change in his pain.

Another ECG was recorded shortly after the first and was identical.

The initial troponin I returned at 0.018 ng/mL (normal up to 0.030).

The second trop returned at 0.28 ng/mL (positive).  The patient's pain had not entirely resolved, but was at 3/10.

Persistent pain with objective evidence of MI is reason for aggressive management.

Another ECG was recorded:

There are some subtle, nondiagnostic differences in the inferior T-waves, which are now smaller.  There are no unequivocal signs of coronary occlusion.  The ST segments and T-waves are dynamic.

He was started on antiplatelet and antithrombotic therapy.

Suddenly, the patient developed sinus brady at a rate of 30, and had a 4 beat run of VT.  He was hypotensive.  The bradycardia and hypotension responded to atropine.

A bedside echo showed probable inferior wall motion abnormality.

Another ECG was recorded:

Sinus bradycardia.  Now the ST depression and T-wave inversion in aVL is resolved.  The inferior ST segments are completely isoelectric.  The inferior T-waves are smaller.  There is new T-wave inversion in III and aVF.

These are all signs of reperfusion, and indicate that the subtle findings of the first ECGs (minimal ST elevation, T-waves, ST depression and T-wave inversion in aVL) were all due to ischemia.

He was taken to the cath lab and found to have a severe 90% thrombotic lesion of the proximal to mid RCA.

In our study of inferior STEMI vs. non-MI causes of inferior ST elevation, any ST depression in aVL was almost perfectly diagnostic (sensitive and specific) for inferior MI.

Awareness of this finding allowed for vigilant observation and serial ECGs even in the absence of a typical history.

Dynamic T-wave inversion (apparent Wellens' waves), all troponins negative: Unstable Angina

This middle-aged woman presented with increasing intermittent substernal chest discomfort similar to her GERD, but not relieved by the usual therapies.  She was given an aspirin.  She had the following ECG recorded in the ED:

A very astute physician read this as "biphasic T-waves in V3 and V4."  There is also T-wave inversion in aVL.  This is very suggestive of Wellens' syndrome with a proximal LAD lesion.

A subsequent ECG was recorded:

Not much changed

The patient was admitted to observation.  Her troponins [Ortho Clinical Diagnostics, Limit of detection is 0.012 mcg/L, 99% reference value ("positive" troponin) of 0.034 mcg/L] were less than 0.012, then 0.015, then less than 0.012.  

The ECG findings were not commented upon by the inpatient team, and the patient technically "ruled out" for MI.  After a careful evaluation that did not suggest an ischemic etiology, she was sent home without a stress test and with a diagnosis of "reflux."

2 weeks later, the patient presented with the same symptoms, happening 5 times between 6 AM and noon, never lasting longer than 15-20 minutes.  Here was here initial ECG with pain:

This time there are full blown Wellens'-like T-waves in V2-V5, I, and aVL, nearly diagnostic of a proximal LAD stenosis.  

Then the pain resolved 25 minutes later, and this ECG was recorded:

There are PVCs, but the Wellens' T-waves have resolved.  This is typical of unstable angina: when there is infarction, the T-waves will evolve by becoming deeper and more symmetric over many hours' time.  See link below.

She was started her on heparin and eptifibatide.  105 minutes later (it is uncertain whether the patient had another episode of pain that she did not report). 

Wellens' waves are back

The next day at 7 AM this was recorded:

Wellens' T-waves are again less prominent

Troponins never became "positive:"  The first level was "normal," the second was "normal," then 3rd was 0.021, 4th 0.029, 5th 0.032, never climbing above the 99% reference value of 0.034 mcg/L.

Such a rise under the 99th percentile is highly suggestive of, but not diagnostic of, ischemia.  With contemporary troponins, the 10% coefficient of variation (10% CV) is not low enough to be certain that these are real differences, though they probably are.  High sensitivity troponins have very low 10% CV and can accurately measure changes below the 99% level. They may help diagnose unstable angina in the future.

Case continued

Later that day, the patient underwent an angiogram and had a 95% stenosis of the proximal LAD with thrombus, and another of the first diagonal off the LAD.  Both were stented.

Comment

This is NOT Wellens' syndrome.

In Wellens' syndrome, the T-waves are inverted when the patient is pain-free.  The underlying pathology is that during the pain, the artery was occluded and there was (unrecorded) ST elevation.  After reperfusion, when the patient is pain-free, if there is some infarction, the terminal part of the T-waves invert (Wellens' Pattern A).  Over time, the T-wave inversion evolves to become deeper and more symmetric (Wellens' Pattern B).

This case is distinct from Wellens' syndrome.  This ischemic T-wave inversion is during the most ischemic phase but without complete occlusion and ST elevation. The T-waves normalize when the ischemia resolves.  There is no evolution of T-wave inversion because there was little or no infarction.

Here is typical Wellens' evolution:

http://hqmeded-ecg.blogspot.com/2011/03/classic-evolution-of-wellens-t-waves.html

When these T-waves normalize, it is NOT pseudonormalization:

In this case, the T-waves were dynamic, inverting, then normalizing.  One might be tempted to call these normalizing T-waves "pseudonormalization."  But this term is the name for the becoming-upright of a T-wave when the artery is re-occluding.  See these posts:
http://hqmeded-ecg.blogspot.com/search/label/pseudonormalization

In this case, the ischemia is resolving without significant infarction, so that the T-waves truly normalize.

1) Even ACS with negative troponins may be strongly suspected by ECG analysis.
2) Troponin rise and fall, even below the 99th percentile, strongly suggests ACS
3) Dynamic T-waves are an infrequent but potentially important sign of ACS.

Long QT: Do not trust the computerized QT interval when the QT is long

A middle-aged male with h/o DM was found down.  He was hypoglycemic (mild, 45 mg/dl) and had pneumonia with hypoxia.  He had this ECG recorded:

Sinus rhythm with slight right axis deviation and non-diagnostic T-wave inversions

He received azithromycin and ceftriaxone for community acquired pneumonia.  Then he became very agitated in spite of correction of hypoglycemia, and was given a total of 15 mg of haloperidol.  His K returned at 2.8 mEq/L and ionized Calcium at 3.82 mEq/L (normal, 4.4 -5.2).  Magnesium was 1.5 mEq/L (normal, 1.3 - 2.0)

A troponin returned slightly elevated, so another ECG was recorded:

There is now sinus bradycardia with a bizarrely long QT interval.  The computer read the QT as 450 ms when it is really around 700 ms.

This is a very dangerously long acquired long QT, and is due to a combination of azithromycin, haloperidol, hypokalemia, and hypocalcemia (note the long ST segments which are typical of hypocalcemia).

He was given Magnesium, Calcium, and Potassium, monitored in the ICU, and the eventually the ECG corrected as the drugs metabolized.

An echocardiogram showed no wall motion abnormality and the elevated troponin was not thought to be due to a Type 1 MI in this critically ill patient.

Comment: computer and manual QT measurement

Computer algorithms for measure the QT interval are good for normal QT intervals but not for long QT intervals, and are particularly inaccurate for very long QT intervals.

The QT is the interval from the beginning of the QRS to the end of the T-wave.  It should be the longest QT interval of all 12 leads, and this is the way most computer algorithms measure it.  However, taking the longest of leads II, V2, V5, and V6 will usually do.

One does not actually measure the end of the T-wave: instead, the technique involves drawing a line along the maximum downslope of the T-wave and measuring from where it intersects with the isoelectric line.  Here is a graphic of this from Life in the Fast Lane (same as the graphic in this link).

Once you measure the QT, then the most commonly used heart rate correction is the Bazett formula, which is the QT divided by the square root of the preceding R-R interval (example: if the preceding R-R interval is 810 ms = 0.81 sec, then the square root of 0.81 is 0.9, so a QT interval of 450 ms would result in a QTc of 500 ms (450 divided by 0.9 = 500).

Let's measure the QT on the above tracing:

Lead V5 is the most convenient and appears to be as long as any other lead.  The first vertical line is the start of the QRS.  The angled line is on the steepest part of the T-wave slope.  The horizontal line is the isoelectric line.  The intersection of the angled line and the isoelectric is where the second vertical line is.  Then the measurement is between the two vertical lines.  There are 3 large 5 mm boxes (each 200 ms) + 2.25 mm to the left and 1.25 mm to the right.  That is 18.5 mm x 40 ms per mm = 740 ms.  QTc = 740 divided by the square root of the R-R interval of 1.08 seconds = 740 divided by 1.04 = 711 ms.
QTc = 711 ms

Life in the Fast Lane has an excellent overview of QT prolongation.

Acquired long QT, and how it causes torsade: It is usually due to drugs.  The list is long.  And also due to electrolyte abnormalities, especially hypoK and hypoMg.   Corrected QT interval (Bazett correction = QT divided by the square root of the preceding R-R interval in milliseconds) is usually great than 600 ms.  Torsades in acquired long QT is much more likely in bradycardia because the QT interval following a long pause is longer still.  Thus, torsades in acquired long QT is called "pause dependent": if there is a sinus beat after a long pause (which creates a longer QT interval), then an early PVC ("early afterdepolarization," EAD) is much more likely to occur during repolarization and to initiate torsades.  The usual sequence is: sinus beat, then early PVC, then a long pause because the PVC was early, which then results in a particularly long QT, then another PVC with "R on T" that initiates torsades.

See this case of Polymorphic Ventricular Tachycardia.