Wednesday, July 1, 2020

This patient has a severe electrolyte abnormality. Can you tell what it is? (Not hypokalemia)

This patient has a severe electrolyte abnormality.  Can you tell from the ECG what it is?
IMPRESSION by computer and physician overread:
Heart Rate 120
Prolonged QT

P-R Interval 220 ms
QRS Interval 84 ms
QT Interval 349 ms 
QTC Interval 419 ms
P Axis 125
QRS Axis 13
T Wave Axis 2

I think the computer got the QT wrong.  It is at least 360 ms and possibly as high as 400 ms.
Hodges Corrections using 360 ms: QTc = 465 ms

Smith Impression: The ECG shows sinus tachycardia, some prolongation of the QT interval, and ST depression.  None of these findings are specific to this patient's severe hypernatremia.  With ST depression, one must entertain hypokalemia as well as ischemia, but the K was normal.

Clinical History and Lab data:

This patient complained of diarrhea, and attempted to treat his diarrhea with homemade electrolyte solution.  He presented with altered mental status and his Na = 175 mmol/L.  Other electrolytes were normal.

His troponin was modestly elevated and the ST depression is probably due to demand ischemia.

I don't believe you can make the diagnosis from the ECG.

None of the findings on the ECG are attributable to hypernatremia.  As far as I can tell, there are no typical ECG findings in the context of severe hypernatremia.  One might guess that high extracellular Na would result in more rapid influx of Na during depolarization, and therefore a narrower QRS.  On the other hand, it might also result in hyperpolarization of the membrane and thus slower depolarization.  As for repolarization, one might make arguments both ways.  I have not seen any experimental data nor heard arguments from experts who would know much more.  In any case, the clinical data is scant.

Here the QRS is 84 ms.  While this is on the low end of normal.  No ECG was recorded after normalization of the Na, but there was a prior ECG available and the QRS duration was 92 ms.  So perhaps hypernatremia shortens the QRS a bit, but that is really conjecture at this point.

There is this unfortunate case report, with a review:
"ECG findings in the setting of extreme hypernatremia are rare and out of date. Levine in 1955 described a case of accidental sodium ingestion that resulted in nonspecific T-wave inversions [3]. ECG changes in dogs after rapidly inducing hypernatremia have also been studied; findings showed QT prolongation and PR interval shortening (as was seen in our patient) in addition to decreased P and QRS amplitude [4]"

Another case here (unfortunately, although they show all 4 ECGs, they are too small, with low resolution, to make conclusions.
Extreme hypernatremia (254 mmol/L) and electrocardiogram findings
In this case, the extreme hypernatremia on Day 1 was associated with sinus tachycardia at a rate of 104 bpm, short PR interval of 88 ms, diffuse ST depression, and QTc 428 ms. Day 3 ECG with sodium of 202 mmol/L. ST depression resolved and the PR interval continued to shorten, HR 63 bpm, QTc, 489 ms. Day 5 ECG with sodium of 184 mmol/L. Minimal T-wave inversion present in V2–V6. Day 16 ECG with sodium of 140 mmol/L, sinus tachycardia otherwise no other abnormalities

MY Comment by KEN GRAUER, MD (7/1/2020):
The subject of today’s case relates to a particular electrolyte abnormality. We need to acknowledge at the outset, that other than for hyperkalemia — an “electrolyte disorder” will rarely be the primary indication for obtaining an ECG. That said — the ECG may provide invaluable assistance in assessing some patients with certain types of electrolyte disorders, especially for explaining some of the ST-T wave abnormalities that may be seen.

KEY Point: An important concept to appreciate, is that some electrolyte disorders can (and should) be recognized immediately (ie, before blood test confirmation returns).
  • For example — numerous cases of life-threatening hyperkalemia have been published in Dr. Smith’s ECG Blog, in which IV calcium gluconate is immediately indicated based on the ECG, even though the precise serum K+ level is not yet known (See the September 8, 2019 post in Dr. Smith’s ECG Blog, among many others).
  • An important clinical clue to the likelihood of there being an electrolyte disorder — is the History. Knowing that a patient has severe renal impairment; is markedly dehydrated; is prone to presenting in DKA; or that the patient has been vomiting all day prior to coming to the ED — serve as invaluable clues that one or more serum electrolyte disorders may be present.

That said — many electrolyte disorders are not instantly recognizable from the ECG. This is the situation in today’s case. For clarity — I’ve reproduced the presenting ECG from today’s case in Figure-1.
  • WHAT are the ECG findings in this tracing that are consistent with an electrolyte disorder?
  • Can you tell WHICH electrolyte disorder is suggested by this tracing?

Figure-1: The initial ECG in today’s case (See text).

My THOUGHTS on ECG #1: Although the history in today’s case should clearly make you suspicious of electrolyte disturbance (ie, The patient presented with altered mental status, after trying to treat his diarrhea at home with a self-made electrolyte solution) — I did not think his initial ECG was diagnostic.
  • There is much baseline artifact. This renders P wave morphology to be highly variable — although, as per the RED arrows in lead II — I thought the rhythm in ECG #1 was identifiable as sinus tachycardia at a regular rate of ~118/minute. The PR interval looked normal, given this rapid rate (ie, the PR interval often shortens a bit with tachycardia). The QRS complex was narrow.

Assessment of the QTc interval for ECG #1 is worthy of discussion. In my experience — the clinical utility of assessing the QTc at faster heart rates is limited.
  • The MD Calculator web site offers automatic calculation of the QTc (Corrected-for-Rate) for your choice of the 4 most commonly cited calculation methods. All one does to use this calculator, is enter the heart rate and the QT interval you measure (in terms of the number of small boxes contained within the longest QT interval that you measure).
  • Using the calculator at this site — I entered the heart rate in ECG #1 ( = 118/minute) — and the longest QT interval I measured ( 350 msec, which corresponds to 350 ÷ 40 msec/small box = 8.75 small boxes). Plugging these 2 numbers into the various formulas on the MD Calculator site resulted in an estimated QTc = 491 msec (for the Bazett formula)  439 msec (for Fridericia) — 426 msec (for Framingham) — and 452 msec (for Hodges).
  • Whereas the calculated value for the Bazett-measured QTc in ECG #1 would be prolonged (ie, 491 msec) — and the Hodges-measured QTc would be borderline (ie, 452 msec) — the other 2 formulas yielded a QTc within the normal range.
  • NOTE: This discrepancy in estimated QTc values highlights the variation between the various methods, especially at faster heart rates. In particular — the Bazett method is notorious for over-estimating the QTc at faster heart rates (For more on this subject — SEE the detailed discussion by Dr. Smith in the November 24, 2017 post).
  • BOTTOM Line regarding the QTc in ECG #1: I fully acknowledge my uncertainty regarding assessment of the QTc in ECG #1. Subjectively, the QTc “looks” to be at least slightly prolonged (ie, it clearly measures more than half the R-R interval) — but given the degree of tachycardia, I feel the QTc in ECG #1 is of limited clinical utility.

Returning to Systematic Assessment of ECG #1:
  • The frontal plane axis is normal (about +20 degrees).
  • There is no chamber enlargement.
  • Regarding Q-R-S-T Changes: There are small, narrow q waves in lateral leads (I not believe there is a Q wave in lead III, but instead an rSr’ complex in ths lead) — R wave progression is normal (transition occurs between leads V2-to-V3) — and, there are diffuse nonspecific ST-T wave abnormalities (some ST-T wave flattening, with 1 mm of ST depression in lead V3).

MY Clinical Impression of ECG #1:
  • The main findings in ECG #1 are sinus tachycardia with nonspecific ST-T wave abnormalities. The QTc looks like it may be prolonged, though this is less certain given the rapid rate.
  • IF this patient would have been having worrisome chest pain — the ST-T wave changes could be consistent with ischemia, that might be acute. Clinical correlation (and serial tracings, as well as comparison with prior tracings) would be needed to sort this out.
  • However, in view of the history we were given (ie, This patient presented with altered mental status, after trying to treat his diarrhea at home with a self-made electrolyte solution) — I suspected the patient might have hypokalemia and/or hypomagnesemia.
  • The above said — I never would have guessed from ECG #1 that this patient’s serum sodium was dangerously elevated to 175 mmol/L.

Serum Sodium and the ECG:
  • Severe hypernatremia is not a common disorder. As a result — the literature on ECG findings with severe hypernatremia is scarce. Dr. Smith cites the reference by Arambewela et al (J Med Case Rep, 2016) — in which a case study on this subject showed ECG changes of sinus tachycardia; PR interval shortening (though PR interval lengthening is often difficult to assess at faster heart rates); QT interval prolongation; and potential for precipitating malignant arrhythmias.
  • NOTE: Both hypo- and hypernatremia may be associated with other electrolyte disorders, as well as with other conditions associated with increased morbidity and mortality. My clinical experience (and what I was able to glean from the scant literature available) — left me with the distinct impression that while an ECG such as that seen in Figure-1 could be consistent with severe hypernatremia — diagnosis of a specific serum sodium electrolyte disorder is unlikely to be forthcoming from assessment of an ECG, especially in the absence of a suggestive history.

In the interest of honing ECG interpretation skills — Consider the following 2 ECGs that are shown in Figure-2 (Both of these tracings are taken from the February 19, 2015 post in Dr. Smith’s ECG Blog).
  • WHICH electrolyte disorder(s) is suggested by the findings in ECG #2, and in ECG #3?

Figure-2: Two ECGs that I’ve selected from patients with electrolyte disorders. WHICH electrolyte disorder(s) is suggested in each tracing? HINT: The inserts of leads V2 and V3 for ECG #3 show a QRST complex in these leads after treatment (See text).

The rhythm in both tracings shown in Figure-2 is sinus, albeit with bradycardia (as well as with lots of artifact) for ECG #2. That said — the principal abnormal findings relate to assessment of ST-T wave changes:
  • ECG #2 — The QTc is clearly prolonged. At the end of a relatively normal (flat) ST segment — is seen an otherwise seemingly unaffected T wave. Of note — the T wave in several leads in ECG #2 is more-peaked-than-is-normally expected (ie, in leads I, II, V3-thru-V6). It should be emphasized that although the T wave in several of these leads is not overly tall — it is more peaked than expected. This is the tent sign” at the end of the desert (ie, after a normal but prolonged ST segment is seen an “all alone” T wave)Hypocalcemia often accompanies hyperkalemia — so the T wave appearing at the end of a long ST segment is often peaked in renal failure patients with both abnormalities.

ECG #3 — The history in this case is insightful. My years in primary care taught me to be leery of older patients presenting with an unusual history of atypical pain in some specific part of their body. While clearly patients with “frozen shoulder” may experience exacerbations in the degree of their shoulder pain — I immediately became suspicious after hearing this history and seeing ECG #3, that the patient had hypercalcemia secondary to malignancy. The remarkable ECG finding in this tracing is a short QTc interval.
  • PEARL #1 (re Hypercalcemia): While the textbook description of ECG findings with hypercalcemia is simply “QT interval shortening” — QT shortening is not an easy ECG finding to recognize (even when you are looking for it). In addition, what is not described in textbooks — is how high the serum Ca++ must go before such QT interval shortening occurs. Over my 3 decades as family medicine attending (working in and out of the hospital) — I religiously scrutinized the ECGs of all patients I encountered in whom serum calcium levels were elevated. In my experience — NO change in ECG appearance was noted in the overwhelming majority of hypercalcemic patients until their serum Ca++ was significantly elevated (ie, generally over 12 mg/dL). The patient whose tracing is shown in ECG #3 had a serum Ca++ = 15 mg/dL. He was found to have lung cancer with shoulder metastases accounting for the increase in his extremity pain.
  • PEARL #2 (re Hypercalcemia): More than simply QT interval “shortening” — the principal ECG finding of significant hypercalcemia is a short-Q-to-peak-of-T interval. By this I mean that the time it takes for the T wave to attain its peak is shortened with significant hypercalcemia. I know of no measurement to quantify this shortened time-until-T-wave-peak. Instead — it is a subjective judgment — that with experience (armed by an increased index of suspicion for the case-at-hand) YOU can appreciate. In Figure-2 — the vertical BLUE lines in leads V2 and V3 of ECG #3 are placed over the peak of the T waves in these leads. Doesn’t the time until attaining this T wave peak “look” short? Now look at the Inserts in leads V2 and V3 for ECG #3. I’ve placed one QRST complex within each insert from the repeat ECG after correction of the elevated serum Ca++ level. Doesn’t the time until T wave peaking (marked by vertical GREEN lines within the inserts) now look longer (and more normalafter correction of the serum Ca++ level?

Concluding NOTE: We’ve featured numerous cases of hyper- and hypokalemia in prior blog posts. Simply enter either hyperkalemia or hypokalemia into the Search Bar on Dr. Smith’s ECG Blog if interested in finding ECG examples of these electrolyte disorders.
  • I’ll conclude today’s post with reference to the January 26, 2020 post (for a sequential Table in My Comment on ECG findings with Hyperkalemia) — and to the May 9, 2020 post (for a sequential Table in My Comment on ECG findings with Hypokalemia).

No comments:

Post a Comment

DEAR READER: We welcome your Comments! Unfortunately — due to a recent marked increase in SPAM — we have had to restrict commenting to Users with a GOOGLE Account. If you do not yet have a Google account — it should not take long to register. Comments give US feedback on how well Dr. Smith’s ECG Blog is addressing your needs — and they help to clarify concepts of interest to all readers. THANK YOU for your continued support!

Recommended Resources