Wednesday, January 7, 2015

Persistent Juvenile T-wave Pattern

This article is written by Brooks Walsh, MD, an emergency physician, as well as Steve Smith, and with help from Ken Grauer, who is quite an ECG whiz.  

Brooks tackles the difficult issue of Persistent Juvenile T-waves (PJTWP). These are slightly asymmetrically inverted T-waves in V1-V3, but not beyond.  The bottom line is that there is little firm guidance on the topic.  


PJTWP important considerations:

1. Patients are typically African American women under age 30.  It is rare in males over 19 years of age to have T-wave inversion beyond lead V1, unless there is lead misplacement or also possibly deep inspiration during recording (1).

2. T-waves are slightly asymmetrically inverted in V1-V3.  T-wave inversion that extends out to V4 and beyond should only be seen in patients under age 12.

3. "Benign T-wave Inversion" is a different form of non-pathologic T-wave inversion.  It does often extend out to V4 and beyond, has some ST elevation, and biphasic T-waves.  It is seen primarily in young African American males.

4. There are no structural cardiac abnormalities.

5. The primary life-threatening pathologies on the differential diagnosis are 
         a) Anterior ischemia (from pulmonary embolism or ACS)
         b) ARVD, Arrhythmogenic Right Ventricular Dysplasia (ARV Cardiomypathy).  It is relatively rare, but causes deadly dysrhythmias.

ARVD: T-wave inversion in V1-V3, with the typical PJTWP morphology, but associated with
 i)  Syncope without a prodrome, 
ii)  PVCs with an LBBB morphology, 
iii) Ventricular dysrhythmias, or 
iv)  Epsilon waves of course are very specific but insensitive for ARVD
v)   Males over age 19, definitely need further evaluation.

6.  Although it is called "persistent", these T-waves may not always be persistent.  Instead, like all benign findings, including early repolarization, it seems that they may be absent on a previous ECG and still be benign.

Persistent juvenile T wave pattern (PJTWP) – persistent confusion?

A 32-year African-American female came to the ED complaining of episodes of palpitations and a “racing” heart. She had a history of DM and HTN. Vital signs and the physical exam were unremarkable. An ECG was obtained:
Figure 1.  There are assymetric T-wave inversions in V1-V3.  Are these normal variants?  Persistent Juvenile T-waves?  Are these pathologic?  Does she have "anterior" ischemia?
This was compared with an ECG recorded 7 years prior:
Figure 2. This previous ECG also has TW inversions in V1 and V2 and a biphasic TW in V3.  
Does the ECG demonstrate a PJTWP?  Does the fact that it was not fully present before preclude the diagnosis of PJTW?
This is a difficult question to answer, but there have been a number of publications in the last few years that shed some light on PJTWP. I’ll review 5 issues that this literature highlights, then circle back to our patient.

Issue 1. Definition of PJTWP
A clear description of the PJTWP is surprisingly difficult to find.

Defining true juvenile T wave patterns
It is worth revisiting the “true” juvenile T wave pattern. Recall that the RV of the neonate has spent 9 months fighting the high-resistance pulmonary circulation, and so the RV is (non-pathologically) hypertrophied. As a result, there may be ECG findings of right ventricular dominance, including T-wave inversion (TWI) in leads V1-V3 or V4 in young children. Generally, this pattern evolves to the adult pattern (i.e. TWI limited to V1) by about 10 years of age.

Characteristics of “true” juvenile T wave pattern include shallow inversions, limited to V1-V3/V4, an asymmetric morphology of the inverted T wave, and no significant ST segment deviation. For example, here is the ECG of a healthy 3 year-old female:

Figure 3. We call these slightly asymmetric T-waves in V1-V3.
Some would call these symmetric, in contrast to the very asymmetric T-waves of, for instance, Left Ventricular Hypertrophy below.

Figure 4. These are T-wave inversions that everyone would call asymmetric

For comparison, here are the symmetric T-waves of Wellens' Pattern B syndrome:
Figure 5. Note the near perfect symmetry of V2 and V3. This is NOT normal, not PJTWP.

The ECG of another healthy 3 year-old, taken from Chan et al. (2)
Figure 6. Notice the inverted T-waves in V1-V3 are slightly asymmetric.

An example of a juvenile T wave pattern in a healthy 11 year-old male is provided in an article by Sharieff and Rao:(3)
Figure 7.  Here the T inversion is limited to V1 and V2; it is slightly asymmetric.

Defining “persistent” juvenile T wave pattern in adults
There are no consistent definitions of this adult variant of TWI. While one author proposed “asymmetric T-wave inversions in right precordial leads, without any other abnormalities”(4) as criteria, not all researchers agree.

For example, at least three articles suggest that PJTWP is typically associated with significant ST segment elevation in those same leads.(5, 6, 7)     

Look at these closely, as we do not agree that this is PJTWP!

Figure 8. One complex of domed ST elevation preceding the TWI.  This is not PJTWP, rather it is Benign T-wave Inversion (BTWI), also known as ST-T Normal Variant (STTNV) (8).

Choo 2002
Figure 9.  Domed ST elevation preceding TWI -- we do not believe this is PJTWP, rather it is BTWI, the other normal variant

2009 Papadakis
Figure 10.  Domed ST elevation preceding TWI -- we do not believe this is PJTWP, rather it is BTWI, the other normal variant

All three of these ECGs show domed ST Elevation that precedes the TWI in the precordial leads, a feature that is not usually seen in children. This pattern has been termed “benign T wave inversion” (BTWI) or “ST Elevation and Inverted T Wave” or ST-T Normal Variant (STTNV) by various authors.   See numerous examples of BTWI here.  In contrast to PJTWP, this STE/TWI pattern of BTWI is found more often in males than females,(8) and is considered by some to be a training-related variant.  It is especially common in African American males.(8)

Issue 2. PJWTP is found more often in women.
Most studies show that anterior TWI is found more often in women than men. In a Finnish study, the distinct majority (87%) of the people with right-precordial TWI were women (9), and a retrospective review done in New Jersey also found a similar proportion.(4). On the other side of the world, in a population of Israeli Bedouins, only women showed this pattern (10).

Issue 3. TWI (PJTWP or BTWI?) is found more often in people of African heritage

Similar to other atypical patterns of repolarization abnormalities, PJTWP appears to be seen more common in patients of African heritage. In a cohort of black and white females in the UK, 15% of the black females manifested TWI in anterior leads, while only 4% of the white females did.(11).  Similar results were seen in a cohort of British and French athletes.(12).  In a group professional American football players, 4.3% of the black players showed this pattern, while only 1% of the white players did.(6)  A 2008 study found that TWIs (of unspecified location) were far more common in black athletes than white.(13)   Unfortunately, these studies are complicated by the confusion between PJTWP and BTWI patterns.

Issue 4. It may not be part of the “athlete’s ECG.”
A number of ECG variants have been described in highly trained athletes; e.g. low-grade AV blocks, pseudo-LVH patterns, RSr', and early repolarization.(14) It is unclear, however, if anterior TWI is part of this group of variants.

A number of studies have suggested that anterior T wave inversions are more common in athletes, and that they resolve with cessation of intense training.(15)  However, Sharma found an equal incidence of anterior TWI  greater than 2 mm in both athletes and non-athletes.(16)

Other experts agree with this perspective.(17) Accordingly, at least 3 different groups have recommended that athletes who have TWI in V2 and V3 should receive further evaluation, even if currently asymptomatic.(5, 18, 19)

Issue 5.  PJTWP is considered after ischemia, PE, and ARVC have been excluded.
A diagnosis of PJTWP should be arrived at only after consideration of more dangerous causes of anterior TWI. Such ECG changes could reflect severe COPD, PE, or pulmonary hypertension. Posterior MI or anterior ischemia should also be ruled-out.

Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a rare disease, with ECG manifestations that could be mistaken for PJTWP.  Criteria for recognizing ARVC on the standard ECG include “inverted T waves in right precordial leads (V1, V2, and V3) or beyond in individuals greater than 14 years of age (in the absence of complete right bundle-branch block QRS ≥120 ms)” as a major criterion for diagnosis.”(20) Clearly, in the right context such as syncope, palpitations, or tachycardia, ARVC must be considered before diagnosing PJTWP on the ECG.  

Here are diagnostic criteria for ARVD from Eur Heart Journal Task Force (full text link): Diagnosis of arrhythmogenic right ventricular cardiomyopathy/dysplasia: proposed modification of the task force criteria.

So, does our patient have persistent juvenile T wave pattern?
The 2014 ECG shows asymmetric T wave inversion in leads V1 – V3, without ST segment elevation or other concerning findings on the ECG. A review of her old ECGs showed, however, that this TWI was not unchanged from prior, and was more pronounced than 7 years ago. In particular, the T wave in V3 is now over 2 mm deep.

A cardiology consultation was obtained, serial troponin levels were negative, and an echocardiogram from 3 years prior was found to be normal. She was discharged from the ED with plans for outpatient follow-up with cardiology.
Figure 11.  Note that the T-wave inversions in 2014 are deeper than in 2013.  Can we prove that this is still normal?  Or Abnormal?

Multiple ECGs were obtained in each patient, so that the chance of a technical error (lead placement) causing this pattern is unlikely.  In case #1, 2 ECGs were obtained in different months of 2007 and 2014 and were consistent.

One may object that without definitive evaluation using echocardiography, angiography, MRI, etc., that we cannot be certain that the TWI is not due to an undiagnosed structural disorder, including ARVD.  Evaluation was pursued only to the degree that the indivudual clinician felt was warranted for the presenting complaint.  

On the other hand, we are not aware of any longitudinal studies of normal populations which confirm that what appears to be PJTWP does NOT develop later.  We do know that many T-wave inversion patterns are benign.

Given that this pattern is commonly presumed to be benign, clinicians may have "underinvestigated" the ECG findings in this case.  Biases about TWI in female African American patients may play a role in limited investigation, leading to premature diagnostic closure.  

These limitations argue for reconsidering the benignity of PJTWP.

So – can you diagnose PJTWP if the pattern is not, in fact, persistent? Despite the number of new articles on the subject, there is no guidance here.

Furthermore, as discussed in Issue #1 above, much of the literature regarding PJTWP includes ECGs with significant ST elevation in the anterior leads, a distinctly unjuvenile pattern. How distinct is this STE/TWI pattern from “true” PJTWP? Is it a minor variant, or is it clinically important? Again, the answer isn’t clear from the recent results.

I guess you could say that our case and review suggest that “persistent juvenile” T wave pattern may be neither persistent nor juvenile.

1.            Marcus FI.  Prevalence of T-Wave Inversion Beyond V1 in Young Normal Individuals and Usefulness for the Diagnosis of Arrhythmogenic Right Ventricular Cardiomyopathy/Dysplasia.  Am J Cardiol 2005;95:1070-1071.
2.            Chan TC, Sharieff GQ, Brady WJ. Electrocardiographic Manifestations: Pediatric ECG. J. Emerg. Med. 2008;35(4):421-430. doi:10.1016/j.jemermed.2007.09.039.
3.            Sharieff GQ, Rao SO. The Pediatric ECG. Emerg. Med. Clin. North Am. 2006;24(1):195-208. doi:10.1016/j.emc.2005.08.014.
4.            Kaid KA, Maqsood A, Cohen M, Rothfeld E. Further characterization of the “persistent juvenile T-wave pattern” in adults. J. Electrocardiol. 2008;41(6):644-645. doi:10.1016/j.jelectrocard.2008.08.028.
5.            Uberoi A, Stein R, Perez MV, et al. Interpretation of the Electrocardiogram of Young Athletes. Circulation 2011;124(6):746-757. doi:10.1161/CIRCULATIONAHA.110.013078.
6.            Choo JK, Abernethy III WB, Hutter Jr. AM. Electrocardiographic observations in professional football players. Am. J. Cardiol. 2002;90(2):198-200. doi:10.1016/S0002-9149(02)02454-2.
7.            Papadakis M, Basavarajaiah S, Rawlins J, et al. Prevalence and significance of T-wave inversions in predominantly Caucasian adolescent athletes. Eur. Heart J. 2009;30(14):1728-1735. doi:10.1093/eurheartj/ehp164.
8.            Roukoz H.  Wang K.  ST Elevation and Inverted T Wave as Another Normal Variant Mimicking Acute Myocardial Infarction: The Prevalence, Age, Gender, and Racial Distribution.  Annals of Noninvasive Electrocardiology 16(1):64-69, January 2011.   doi:10.1111/j.1542-474X.2010.00410.x.
9.            Aro AL, Anttonen O, Tikkanen JT, et al. Prevalence and Prognostic Significance of T-Wave Inversions in Right Precordial Leads of a 12-Lead Electrocardiogram in the Middle-Aged Subjects. Circulation 2012;125(21):2572-2577. doi:10.1161/CIRCULATIONAHA.112.098681.
10.          Assali A-R, Khamaysi N, Birnbaum Y. Juvenile ECG pattern in adult black arabs. J. Electrocardiol. 1997;30(2):87-90. doi:10.1016/S0022-0736(97)80014-3.
11.          Malhotra A, Dhutia H, Gati S, et al. 103 Prevalence and significance of anterior T wave inversion in females. Heart Br. Card. Soc. 2014;100 Suppl 3:A60. doi:10.1136/heartjnl-2014-306118.103.
12.          Rawlins J, Carre F, Kervio G, et al. Ethnic Differences in Physiological Cardiac Adaptation to Intense Physical Exercise in Highly Trained Female Athletes. Circulation 2010;121(9):1078-1085. doi:10.1161/CIRCULATIONAHA.109.917211.
13.          Magalski A, Maron BJ, Main ML, et al. Relation of Race to Electrocardiographic Patterns in Elite American Football Players. J. Am. Coll. Cardiol. 2008;51(23):2250-2255. doi:10.1016/j.jacc.2008.01.065.
14.          Wu J, Stork TL, Perron AD, Brady WJ. The athlete’s electrocardiogram. Am. J. Emerg. Med. 2006;24(1):77-86. doi:10.1016/j.ajem.2005.04.009.
15.          Wilson MG, Sharma S, CarrĂ© F, et al. Significance of deep T-wave inversions in asymptomatic athletes with normal cardiovascular examinations: practical solutions for managing the diagnostic conundrum. Br. J. Sports Med. 2012;46(Suppl 1):i51-i58. doi:10.1136/bjsports-2011-090838.
16.          Sharma S, Whyte G, Elliott P, et al. Electrocardiographic changes in 1000 highly trained junior elite athletes. Br. J. Sports Med. 1999;33(5):319-324.
17.          Corrado D, Biffi A, Basso C, Pelliccia A, Thiene G. 12-lead ECG in the athlete: physiological versus pathological abnormalities. Br. J. Sports Med. 2009;43(9):669-676. doi:10.1136/bjsm.2008.054759.
18.          Drezner JA, Ackerman MJ, Anderson J, et al. Electrocardiographic interpretation in athletes: the “Seattle Criteria.” Br. J. Sports Med. 2013;47(3):122-124. doi:10.1136/bjsports-2012-092067.
19.          Corrado D, Pelliccia A, Heidbuchel H, et al. Recommendations for interpretation of 12-lead electrocardiogram in the athlete. Eur. Heart J. 2010;31(2):243-259. doi:10.1093/eurheartj/ehp473.
20.          Marcus FI, McKenna WJ, Sherrill D, et al. Diagnosis of Arrhythmogenic Right Ventricular Cardiomyopathy/Dysplasia Proposed Modification of the Task Force Criteria. Circulation 2010;121(13):1533-1541. doi:10.1161/CIRCULATIONAHA.108.840827.
21.          Dalal D, Nasir K, Bomma C, et al. Arrhythmogenic Right Ventricular Dysplasia A United States Experience. Circulation 2005;112(25):3823-3832. doi:10.1161/CIRCULATIONAHA.105.542266.
22.          Jha AK, Varosy PD, Kanaya AM, et al. Differences in Medical Care and Disease Outcomes Among Black and White Women With Heart Disease. Circulation 2003;108(9):1089-1094. doi:10.1161/01.CIR.0000085994.38132.E5.


  1. Well done discussion by Brooks Walsh (!) - who confronts key issues on distinction between PJTWP vs other potentially more serious conditions. I'll expand on a few points:

    "Symmetry" of T waves is in the eyes of the beholder. Whether one calls a T wave "symmetric" or "asymmetric" depends in large part on what one calls the onset of the "T" wave (vs the end of the ST segment). Best illustration of this concept lies with the 1st and 3rd QRS complexes in lead V6 of Figure 4 - in which the ST segment sags (slow downslope) on its way to a T wave that then quickly returns to the baseline. There is baseline wander in V5 of Figure 4 - but if one uses a late onset for the beginning of the T wave for the middle complex in lead V5 of Fig. 4 - there really is not the typical asymmetry of LV "strain" ... That said - there can be NO DOUBT that the deep T wave inversions in V2,V3,V4 of Figure 5 are clearly symmetric by any definition that one might use. The relevance of this admittedly "picky" point is for each of us to develop our own consistent internal definition for how the T wave inversion in lead V3 of Figure 6 (due to PJTWP) looks different from the ischemic-looking inverted T waves in leads V2,V3,V4 of Figure 5.

    Near the end of Brooks' discussion - he raises the question of whether discharge of this patient from the ED was justified without more "definitive evaluation" including Echo, angiography, MRI, etc. The dilemma of how far to go with one's workup in ruling out all possible pathology is never-ending - especially in emergency medicine. One just doesn't want to miss that "needle in the haystack that might lead to the patient's demise". But in this case - looking for entities such as ARVD is truly looking for a needle in a haystack. ARVD is rare - and the number of false positives one is destined to uncover if full evaluation is done for all patients with anterior T wave inversion is innumerable. Often overlooked is the potential for harm by doing extra work-ups (which will doubtlessly uncover incidentalomas leading to more downstream testing with usually negligible real yield). Additional work-up for the 2014 ECG in this case is simply not needed.

    Finally - I think it important to realize the subtleties of comparing sequential tracings for differences. I completely agree with Brooks that it DOES look like the anterior T wave inversion in this case has increased - but arriving at this conclusion is not as simple as might appear. Note in Figure 11 the difference in QRS amplitude between the 2 tracings - with early transition (between V2-to-V3) in 2007 - whereas the QRS remains isoelectric in V3 in 2014. Lead V2 in 2014 might be misplaced - as its QRS amplitude is inexplicably greater than V3. And returning to Figures 1 and 2 (showing the full 12-leads for 2007 vs 2014) - the frontal plane axis is not the same in the 2 tracings, and the T wave in lead III is negative in 2014 (whereas it wasn't inverted in 2007 ( = S1Q3T3 pattern in 2014). And supporting my theory that there was some lead misplacement in 2014 are the much more abrupt morphology changes compared to the more consistent development of R wave amplitude in V3-thru-V6 seen in Figure 2.

    THANK YOU again Brooks for a superb discussion!

  2. Normally P is biphasic most in V1 .In 2014 , p is gradually moving biphasic to positive through V1 to v6 -not misplAced lead.

    1. Good observation, but I don't think they are different enough to alter the assessment.
      Steve Smith

  3. In your previous case,you said T inversion in III and V1 is almost pathognomonic for PE. Is this case

    1. If I wrote that, then I was mistaken. Can you show me the case where I did so? What I would have meant to write is this: "if the choice is between ACS and PE because of T-wave inversion in V1-V3, then T-wave inversion in III makes it much more likely to be PE." This is from 2 studies by Kosuge et al. Here is one:

    2. Thank you for this great post.I have a couple of questions:
      1.If the pacient with the Figure 1. EKG had chest pain, wouldn't PE be on top of the differential diagnosis?
      (at a glance this EKG looks similar- even if for an expert there are differences)

      2.Would a pacient with figure 10. EKG and recently resolved chest pain be suspected of Wellens syndrome- pattern A ? ( - more to the point how could we be sure if Benign T-wave Inversion is realy benign without coronarography?

    3. Adi,
      Thanks for the great questions.
      There are subtle differences.
      Question 1: the PE you reference has T-wave inversion all the way out to V6, which should never be assumed to be benign. All the cases I have posted of PE here have either T-wave inversion beyond V3, or some ST elevation:

      Question 2: Notice that in figure 10 there is a very tall R-wave AND a J-wave in V4 and V5. These are typical of BTWI and NOT of Wellens'.


    4. Thank you for clearing that up for me.Maybe you should consider talking in a post about the different causes of T wave inversion in the precordial leads( for example PE sometimes looks like Wellens' Syndrome for someone which is less experienced ) and how to differentiate between them.Thank you again for this great learning experience.

  4. Great article and thoroughly researched!

    I have found that comparing the onset of the inverted T wave to the point at which it returns to the baseline is very helpful in determining symmetry vs. asymmetry. If the angles formed by the departure from, and the return to, the baseline are basically the same, then the inverted T is symmetrical. If the angle at the beginning of the inverted T is different than at the end of the T (usually obtuse vs. acute), then it's asymmetrical.

    Thanks again for a great blog!

  5. Thanks for this comprehensive review.
    I can see TWI in leads II and aVF and flat T wave in the first ecg presented that weren't exist 7years ago..Is this significant ? ..And are these tiny q waves in the same 3 leads significant?


    1. KhaleD,
      I don't think those are significant changes.
      Steve Smith

  6. Amazing post! I hope You have more of this stuff for us :)

  7. Thanks for this great post...
    2 questions
    1) The patient is symptomatic here.. The TWIs in anterior and inferior leads plus T wave flattenin in lateral leads of 2014 ECG... doesnt it need Atleast an echo besides just Troponin???

    2) You have mentioned - PJTWI(asymmetrical TWIs in V1-V3) and Benign TWI(ST elevation, and biphasic T-waves in V1-V3 +/-v4 and beyond), but in this case(ECG 2014) there are STds and TWIs in V3 and V4.... Is the STd expected finding of PJTWI?(Not mentioning benign TWIs as it has STe according to you...

  8. From Brooks Walsh:
    1. Another echo in the ED could have been warranted, although cardiology didn't feel so. Given the lack of HCM on the older echo, a structurally overt cause of her palpitations was probably unlikely. That being said, wasn't my patient at that visit, and I probably would performed a point-of-care echo myself! Doubt I would have picked up ARVC though!

    2. The 2014 ECG shows TWI in V1-V4, consistent with PJTWP. The STD in V3 and V4 is seems to be slight, about 0.5 mm at worst. The baseline variation in V3 may make this look worse than it really is.

    To be clear, STD has never been described as characteristic of the juvenile pattern. Significant anterior STD should suggest exclusion of ischemia, PE, etc.


Recommended Resources