Monday, August 15, 2011

Altered Mental Status, possible ingestion. What does the ECG show?

A middle aged man was heard to be falling in his apartment.  He was found very agitated, intermittently screaming (and on presentation was intermittently roaring like a lion).  There was no apparent etiology.  He required 10 mg of droperidol for sedation.  He underwent an ECG as a routine part of the evaluation of possible ingestion:

 What jumps out at you?

Answer: There is sinus rhythm with one PAC.  The notable feature is a very long QT interval.  The computer read this as QT = 492 ms, with QTc = 518 ms.  But when the QT gets very long, computers become inaccurate and you must read it by hand. 

Some recommend reading in lead II or V5/V6.  Some say to measure the longest of the 12 QT intervals.  I tend to measure the longest of the 12 QT intervals on the ECG.  I have not measured them all, but at a glance, it looks like V2 and V3 have the longest, and these typically are the longest.  I have blown them up below:

The end of the T-wave has a small hump (wide arrow) which is probably a U-wave and this can make the end of the T-wave appear to be even more delayed than it is.  The narrow arrow more accurately follows the projection of the T-wave to the baseline (horziontal black line).  Using this at the end of the QT interval, I get 540 ms, with QTc as 540/(square-root of R-R interval) = 540/0.91 = 593 ms
(By my eyeballing it, the QT in V5 is also 540 ms.) 

593 ms is dangerously long and may result in torsade de pointes (polymorphic VT), and could be a result of the droperidol, or of metabolic and electrolyte abnormalities, or to many drugs or even be familial.  In any case, it is unsafe to leave it like this, so we gave 2 grams of Magnesium.  A blood gas revealed a pH of 7.75 (entirely a respiratory alkalosis, drawn before intubation and due to his agitation and hyperventilation), which will also cause long QT and torsade.  K was 3.2, which will also contribute.  By this time, we had him on the ventilator and so we intentionally slowed his ventilations to lower the pH.

Repeat pH was normal and Mg was delivered, and we recorded another ECG:

I calculate the QTc at 488 now, which is safe.

The magnesium level returned at 1.2 (low).

Etiology of long QT in this case: 1) hypo Mg 2) hypoK 3) alkalosis  4) droperidol?   5) other drug (we still do not have a diagnosis and the patient is still unable to tell us what meds he is taking)

1) QTc over 500 ms can be dangerous.  When near 600, it is very dangerous.
2) Give Mg, correct to 2.0.
3) Correct high pH (I once had a patient who required mechanical hypoventilation and intravenous HCl to prevent torsade)
4) Correct low K to 4.5.  Hypokalemia causes a large amount of QT dispersion (see #6)
5) Check Ca (normal in this case).  HypoCa causes long QT by lengthening the ST segment without lengthening the duration of the T-wave.  It is uncommon for hypocalemic long QT to result in torsade because it results in only a small amount of QT dispersion.  It is "homogenously prolonged".
6) both QT interval and QT "dispersion" (a measure of the difference, in ms, between the longest QTc and shortest QTc of the 12-leads) are risk factors for torsade.  A QT dispersion of greater than 60 ms is high risk.

Details on measurement of the QT interval can be found here: free full text article.

One should average 3-5 leads, usually take lead II, or the lead that shows the end of the T-wave best, or the leads with the longest QT (which are usually V2 or V3).  Correction for the heart rate must be done, but tends to overestimate the QT interval at fast heart rates and underestimate at low rates, and so other correction methods have been developed.  The U-wave should generally not be included, especially if large, but if small and not distinguishable from the T-wave, the course of action is unclear.

QT dispersion
--Yelamanchi VP, Molnar J, Ranade V, Somberg JC. Influence of electrolyte
abnormalities on Interlead variability of ventricular repolarization times in 12-
lead ECG. Am J Ther 2001;8:117–122.

--Eryol NK et al.  Effects of Calcium Treatment on QT Interval and QT
Dispersion in Hypocalcemia.  Am J Cardiol 91:750-752; March 15 2003.


Droperidol has a black box warning regarding the QT interval prolongation.  Our department has studied this extensively and found it to be a greatly exaggerated danger.

Here are two abstracts:

QT Lengthening after Parenteral Droperidol Administration

Stephen W Smith, Marc Martel, Michelle Biros, Marsha Zimmerman and Peter Chase
Hennepin County Medical Center: Minneapolis, MN   SAEM, St. Louis 2002
Objectives: Recently, the Food and Drug Administration (FDA) warned of a prolonged QT interval and torsade de pointes as a complication of droperidol (DROP). We sought to determine the frequency of a significantly prolonged QT interval (LQT) in patients who received DROP, and to compare this with the QT interval of patients not receiving DROP. Methods: The EmSTAT electronic patient database was searched from January 1, 1997, through November 30, 2001, for all patients who received DROP. Those who had an electrocardiogram (ECG) ordered at least 30 minutes after administration of DROP were identified. These ECGs were reviewed and the computerized corrected QT intervals (QTc) were recorded. A medication-induced QTc of less than 480 ms is generally considered safe; we defined LQT by QTc greater than 480 ms. Medical records of patients with LQT were further reviewed for previous ECGs, contributing medical conditions, and adverse events. The QTc's of 100 consecutive patients who did not receive DROP were reviewed as controls. Data were analyzed with descriptive statistics and Fisher's exact test. Results: 15,374 patients received 18,020 doses of DROP; 682 had an ECG recorded after DROP, 450 were obtained at least 30 minutes after administration. LQT was found in 17 patients, 1 had left bundle branch block (LBBB), 1 had a paced rhythm, 1 had right bundle branch block (RBBB), resulting in a total of 14 with a normal QRS and LQT (3.1%). Four of these 14 had previously documented LQT not associated with DROP. None had an adverse event related to LQT. Of 100 consecutive patients in the control group, 4 had LQT (4.0%) (p = 0.76) Conclusions: Our study does not support an effect of DROP on the frequency of LQT.

QT Prolongation and Cardiac Arrhythmias Associated with Droperidol Use in Critical Emergency Department Patients

Marc Martel, James Miner, Seth Lashkowitz, Mark Danahy, Joseph Clinton and Michelle Biros
Hennepin County Medical Center: Minneapolis, MN   SAEM Boston 2003
Background: QT prolongation and torsade de pointes (TdP) have been reported as a complication of droperidol (Drop). Objectives: To determine the change in the corrected QT interval (QTc) and the incidence of cardiac arrhythmias in critically ill patients who received Drop. Methods: The medical records of all critical care ED patients from 1/1/1997 to 12/31/2001 were hand searched for those who received Drop and an ECG in the ED. Drop dose, ECG time, QTc intervals, and cardiac rhythm were reviewed. ECGs with atrial fib/flutter, right or left bundle branch block, or paced rhythms were excluded. Data was analyzed in 3 groups, patients with an ECG recorded only before Drop, only after Drop, and those with ECGs both before and after Drop. Data was analyzed using descriptive statistics and chi-squared. Results: 11,583 charts were reviewed, 1172 patients received Drop, and 396 had both an ECG and Drop in the ED. 44 patients were excluded due to abnormal rhythm, bundle branch block, or paced rhythm. 96 patients had an ECG only before Drop (mean 33.3min prior), average dose of 2.75mg, and mean QTc of 435.0ms (95% CI 428.1–441.9ms). 186 patients had an ECG only after Drop (mean 25.9min after), average dose of 3.68mg, and mean QTc of 433.3ms (95% CI 427.8 to 438.8ms). 114 patients had ECGs before and after Drop (mean time 28.2min before, 108.8min after), average dose of 2.21mg, and mean QTc of 435.7ms (95% CI 426.7–444.7ms) and 435.8ms (95% CI 427.5–444.1ms) before and after Drop, respectively. The mean ratio of the QTc before and after Drop is 1.005 (95% CI 0.985–1.025). 2 patients had ventricular arrhythmias in the before Drop group, 3 in the after Drop group, and 4 in the before and after Drop group (p = 0.5). 1 patient had an unrecorded event of TdP with a QTc of 466ms after conversion. Conclusions: We detected no statistical difference in the change of the QTc interval or occurrence of ventricular arrhythmias in critically ill patients who received Drop.


  1. Thanks for this case and an easy to understand explanation about long qt syndrome

  2. Sir, is there U wave immediately after inverted T wave in V1?

    1. It does appear so, but if you look down to V2, V3, and lead II, you see that this apparent U-wave is occurring while the T-wave is still there. So it must be a biphasic T-wave.


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