Wednesday, January 13, 2016

T-wave inversion in a young woman with hyperventilation and carpopedal spasm

This case is another good one contributed by Brooks Walsh, with an excellent review.

A 24 year-old African-American female presented to the ED with nausea and vomiting for 1 day. She also complained of paresthesias and cramping in her hands and wrists, as well as perioral paresthesias. She had no significant  personal or family medical history. Vitals signs and exam were remarkable only for hyperventilation with clear lungs, and bilateral carpopedal spasms.

An ECG was obtained to assess the QTc prior to administering ondansetron:

A bedside focused echocardiogram did not demonstrate overt septal or concentric hypertrophy. Labs showed a mild hypokalemia (3.5 mmol/L), and normal calcium. Metoclopramide was chosen in lieu of ondansetron, and IV fluids given; the potassium was not repleted.

Several hours later, the patient reported that her nausea and paresthesias were much improved, and the ECG was repeated:

She was discharged with an outpatient referral to cardiology.

What the heck was that?!
Her second ECG is classic for so-called “persistent juvenile T-wave pattern” (PJTWP), showing shallow, asymmetric inverted t-waves, limited to the right-precordial leads. Such a pattern is found most often in women of Afro-Caribbean heritage, and is understood to be a benign repolarization variant.

Her first ECG, however, is decidedly not PJTWP. The TWI extends from V1 to V5, and is preceded by domed ST-segment elevation (STE) in V2-V4. It appears similar to a recognized normal variant in male athletes of Afro-Caribbean heritage. (Such an ECG pattern in a female or white athlete, however, requires an echo to rule-out cardiomyopathy!)
But the domed STE and T-wave inversion was dynamic, resolving after her nausea and hyperventilation were relieved. Furthermore, there was QTc prolongation seen in the initial ECG, which resolved as well.
Why did she develop such outsized T-wave inversion? Probably from hyperventilation!

It is well-recognized in the older cardiology literature that hyperventilation can produce T-wave inversions. However, the contemporary literature, and especially that in emergency medicine, does not reflect that. For example, the otherwise comprehensive review by Hayden does not mention hyperventilation.

As early as 1955, it was observed that T-wave inversions seemed to be correlated with episodes of heightened anxiety, observed first in an African-American cohort. Wasserburger explained that this pattern should be considered a “functional cardiac variation...." 

Anterior hyperventilation-induced TWI (HITWI) has been seen in both white and African-American patients, and in both children and adults.

Thomsen found HITWI in 15% of children. Interestingly, these changes became less common in the white children with advancing age, while becoming more common in the African-American children.

From: Thomsen 1967}

Biberman studied a cohort of patients who had previously been found to have anterior TWI with hyperventilation. The ages ranged from widely from 19-71, but no information about ethnicity was given. The patients developed TWI and prolonged QTc with hyperventilation, but no relationship with electrolytes,  pH, or pCO2 was seen.

{Figure from Biberman. Marked anterior TWI is seen after 20 seconds of hyperventilation (HV), and QTc increases from 400 ms to 480 ms.}

Lastly, a few recent studies have demonstrated that hyperventilation can cause transient QTc prolongation in children .

It is unclear if hyperventilation-induced TWI (HITWI) is consistently related to race or age. The 1955 Wasserburger study above only enrolled African-American subjects, with an average age of 33. By contrast, a Greek study that found little anterior HITWI enrolled subjects with an average age in the mid-40s; it is unclear if any patients of Afro-Caribbean heritage were enrolled.

So, likely due to hyperventilation - BUT...
Of course, the dangerous causes of anterior TWI must be considered, such as posterior MI, pulmonary embolus or pulmonary hypertension, or even ARVC. Her history and age, though, suggested these were extraordinarily unlikely.  While LVH or HCM are still possible causes of her unusual ECG, the dramatic resolution does not suggest this. Nonetheless, outpatient evaluation by cardiology was warranted and recommended..


  1. In the interest of covering our bases, I have an alternate explanation:
    Look at the QRS-T morphology in V5 on the first ECG. Now look at the QRS-T morphology in V3 on the second tracing; they are identical. I suspect the dynamic T-wave inversions in this case could instead be due to precordial electrode placement differences between the two tracings. We can't prove or disprove that theory, but I think it's worth keeping high on the differential.

    1. Vince,
      I partly agree. I think V3 is V4, but not V5.

    2. ECG changes due ventilatory alkalosis with electrolyte imbalance, first mild hypokalemia.
      Blood gas analysis would have been useful. I'm sure:-)

    3. Yes, but unnecessary, don't you think?

    4. Corrielle CaldwellJanuary 15, 2016 at 8:28 PM

      I had a similar case a few months ago: A patient of the same demographic, hyperventilating but without carpal spasm. She had TWI in all of the precordial leads and at that time her pH was 7.6 and PCO2 was 18. One hour later with a normal respiratory rate she had PJTWP with pH 7.4 and PCO2 30.

    5. Corielle,
      Interesting. Thanks!
      Steve Smith

  2. I would argue that one finding which, even without a previous ECG, hints to the fact that the ECG was exhibiting hyperventilation-induced T-wave inversion (HI-TWI) is the universal flattening (i.e. trend towards inversion) of T-waves in all leads outside of the precordials (including & except aVR in which the T-wave was flattening but trending towards being upright). I suppose another way to state this would be "very low limb-lead T-wave amplitudes in comparison to those in the precordials".


Recommended Resources