No, doubling the paper speed will not reveal hidden P-waves

Apparently I went to the Rick Bukata School of Titling Articles.

A 22-year-old male presents with agitation and delirium after smoking an unknown substance that an equally unknown person on the street offered him. You note a rapid radial pulse at around 150 bpm and attach him to the cardiac monitor:

Figure 1. Initial rhythm at normal paper speed.

Figure 1. Initial rhythm at normal paper speed (25 mm/s).

Well now we’re in a tough spot. It’s difficult to tell whether Fig. 1 shows sinus tachycardia or some non-sinus narrow-complex tachycardia (we’ll use the colloquial shorthand of “SVT” to include all those other options on the differential, including AVNRT, AVRT, ectopic atrial tachycardia, junctional tachycardia, etc…). If it is indeed sinus tach, then the requisite P-waves must be those upright deflections in II and III and superimposed on the T-waves.

Is there something we could do to see if those really are P-waves buried in the T-waves?

If you’re like me, you were probably taught that it would be a clever move to double the paper speed in a situation like this to separate the P’s from the T’s, revealing the diagnosis of sinus tachycardia. Let’s see what happens when we do that.

Here’s the same rhythm strip run at double-speed:

Figure 2. Double-speed

Figure 2. The same rhythm strip as Fig. 1 but performed at double the paper speed (50 mm/s). [Note: this is only the first 5 seconds of the full 10-second strip. Since the paper’s going twice as fast, a 10-second strip at 50 mm/s ends up twice as long, which is a bit unwieldy for posting.]

We still can’t distinguish P-waves from the T-waves..

Then this must be SVT, not sinus tachycardia!

[Spoiler] But this is sinus tachycardia!

It turns out that changing the paper speed does nothing to reveal waves buried in the ECG. I’ll admit, I used to think it would help too since several instructors suggested it would and I never bothered to try it out. Five years ago I even included it in my first list of pointers for spotting 2:1 atrial flutter—a tip I’ve since redacted. The rationale for why this maneuver’s not helpful has to do with the simple reason why ECG waves end up superimposed in the first place:

There are two events happening at the same time!

A P-wave ends up on top of a T-wave when atrial de-polarization and ventricular re-polarization occur simultaneously. No adjustment to the paper speed is going to change that fluke of timing. It doesn’t matter if the paper if going 10 mm/s or 100 mm/s—if the P-wave is happening at the same time as the T-wave, they’re always going to be superimposed on the ECG.

Don’t believe me?

Let’s look at the normal-speed tracing from Fig. 1 again, but this time I’m going to crop it so that we only see the first half of the strip:

Figure 3. This is identical to Fig. 1 except cropped to show only the first 5 seconds.

Figure 3. This is identical to Fig. 1 except cropped to show only the first 5 seconds.

Now, rather than changing the paper speed, I’m going to stretch the image above by doubling its width.

Figure 4. This is the same image as Fig. 3 except stretch horizontally to double the width.

Figure 4. This is the same image as Fig. 3 but stretch horizontally to double the width.

We can now compare our true double-speed printout (Fig. 2) with this simulated version (Fig. 4). It turns out they are absolutely identical.

Figure 5. The top strip is Fig. 2 (50 mm/s) and the bottom strip Fig. 4 (25 mm/s stretched).

Figure 5. The top strip is Fig. 2 (50 mm/s) and the bottom strip is Fig. 4 (25 mm/s, stretched horizontally).

Another way of understanding this is that all of the information contained on the 50 mm/s strip is already available on the standard 25 mm/s strip. Running the paper at double-speed does nothing to uncover “hidden” complexes since every aspect of the ECG is stretched equally.

What it does succeed at, however, is making ECG interpretation more difficult for providers who don’t usually reading tracings at 50 mm/s (probably 99% of our readers and certainly all of our editors). Outside of the electrophysiology lab and a few countries where 12-lead printouts often include a 50 mm/s rhythm strip, these sorts of tracings are a rarity. And since most of ECG interpretation comes down to pattern-recognition, changing the speed of even simple tracings interrupts all the mental shortcuts we’ve spent years building—forcing experienced providers to approach double-speed tracings with the deliberate, uncertain, and unpracticed eyes of novices.

While it’s possible to read ECG’s that way, it’s not fun.

The Outcome

The trick we do suggest in this situation isn’t that tricky: run a 12-lead.

Figure 5. 12-lead ECG run simultaneously with the prior rhythm strips.

Figure 6. 12-lead ECG run simultaneously with the prior rhythm strips.

V1 in the ECG above confirms that, without a doubt, those are P-waves we are looking at in the T-waves. Their timing raises another issue however: The Bix Rule suggests that whenever you see P-waves positioned halfway between R-waves, you need to worry that there could be hidden P-waves buried within the R-waves as well.

So rather than sinus tachycardia at 140 bpm, could we be dealing with atrial tachycardia at 280 /min with 2:1 conduction?

Figure 7. Blue arrows denote P-waves, while red arrows show where buried P-waves could be hidden as predicted by the Bix Rule.

Figure 7. Blue arrows denote P-waves, while red arrows show where buried P-waves could be hidden as predicted by the Bix Rule.

Thankfully, given a short period of time, in a case like this it’s pretty easy to rule-in sinus tachycardia since it will usually respond to management of the underlying condition. The patient was treated with a small dose of lorazepam and supportive care and a repeat tracing run about 15 minutes later.

Figure 8.

Figure 8. 15 minutes later, it is now clear that we are dealing with sinus tachycardia.

For the curious, I’ve also printed the rhythm from Fig. 8 at double-speed.

Figure 9.

Figure 9. The same strip as Fig. 8, this shows sinus tachycardia at double-speed.

And, finally, the 12-lead.

Figure 10.

Figure 10. This 12-lead, again confirming sinus tachycardia, was performed simultaneously with the rhythm strips in Fig. 8 and 9.

So, the next time someone tries to be clever and identify sinus tachycardia by printing a 50 mm/s rhythm strip, be extra clever and don’t waste your time. There’s no workaround for superimposed complexes and unless you’re in the EP lab, a tracing at standard speed has all the information you could possibly need.


After sharing this on various social media platforms I was glad to see a bunch of comments from readers suggesting that running a Lewis lead rhythm strip might have a role here. The Lewis lead is a relative of V1 (though definitely not identical), and since V1 is our best lead for picking out P-waves on that first 12-lead (Fig. 6), it makes sense that the Lewis lead should offer a similarly useful view of the atria. I didn’t mention that option in my original article for two reasons:

  1. I’m prone to digressions so I’m trying to kept my posts “tighter” these days.
  2. I didn’t think of it.

For more information on how to obtain the Lewis lead and some examples of situations in which it might be helpful I suggest starting with the posts below:



  • Paul Pulleine says:

    Nice article and great way of highlighting information on the ECG. Curious as to whether you use modified leads (such as Lewis Leads etc) to highlight atrial activity in concentrating more emphasis over the Atria?

    Hear from you soon

    • Indeed, my co-editors and I love the Lewis lead! I’ve actually had a half-dozen different people bring up that very topic regarding this case so I’ve added an addendum to the post for readers who might not know what we’re talking about.

  • NICE post Vince! I had not encountered 50mm/second tracings until I started regularly interpreting tracings on the EKG Club with its truly worldwide list of colleagues. I believe 50mm/sec speed for recording is commonly used in Germany (as well as selected other countries) — so good to be aware of what such tracings look like — which you wonderfully explain in your post. My reaction is like yours — despite mentally knowing that we’ve simply doubled the speed, I always feel totally disoriented because of innate pattern recognition that usually tells me what the ECG in front of me shows within seconds. None of that works at 50mm/sec …

    This case you show here is of particular interest for 2 additional reasons: i) On the initial simultaneously-recorded 3-lead rhythm strip with the SVT at 150/minute — IF a P wave is hidden within each preceding T wave — then the PR interval would have to be relatively long. Usually, the PR interval shortens some with tachycardia — so that’s why I initially favored a reentry tachycardia. That said, the specific reason I always like to include sinus tachycardia within my differential diagnosis for a regular SVT when atrial activity is not clearly evident is precisely because sometimes there will be sinus tach with hidden P waves …

    ii) The P waves in ECG #2 are not normal. The PR interval IS long (for a rate of 120/minute) — P waves are tall and peaked in inferior leads, and seem to be notched. There is also significant right axis deviation. Was an Echo ever done?

    Again — NICE case that brings home some important points!

    • Unfortunately no echo was ever done. As happens so often in this cases, the patient felt better a couple of hours later and left against medical advice.

      I’ve only got a couple of these sympathomimetic drug intoxication cases in my scanned collection at the moment (I’m sure I have more in the pile on my desk), but something all three have in common is large, somewhat peaked T-waves (though the other two aren’t quite as dramatic as this case). I wonder if it could be an effect of the synpathomimetics, or maybe just luck of the draw.

      Who knows what kind of history he might have had though? As a person who would smoke whatever some random person on the street handed him, there could be a lot more to his history than we were privy to.

  • Alex says:

    >>I believe 50mm/sec speed for recording is commonly used in Germany (as well as selected other countries) — so good to be aware of what such tracings look like — which you wonderfully explain in your post.

    This speed is VERY commonly used in Russia and other post-Soviet countries.
    PS, in Sweden it’s also common to print out aVR upside down…

  • Jaych Hammington says:

    I was taught this technique to determine whether or not the rhythm is irregular when rate may be masking an a fib with rvr, which I feel is probably much more successful although I have not tried it yet. Granted there are several other ways to determine irregularity but I have encountered a few cases it would be an appropriate assessment tool in my earlier years of pt contacts.

  • Vince..

    An excellent case and very well-presented! Sometimes things that seem like a good idea at first really aren’t. Since the electrocardiograph has no effect on the heart, speeding up the ECG is NOT the same thing as slowing down the heart rate.

    T waves are not normally notched and whenever I see notches in the T waves occurring at a fixed R-P or P-R interval, I immediately assume a P wave is hiding there.

    Another thing to consider is what is happening to the heart when you see a tracing such as this. The patient has an atrial systole that is occurring right on top of a ventricular systole. When this happens, that atrial blood has to go somewhere and forward is often not an option. Look for large pulsations in the neck veins (cannon a waves). Just think what could happen to an elderly patient with COPD and poor cardiac reserves: he/she could get into hemodynamic and respiratory distress very easily due to the back up of blood into the lungs coincident with a decrease in cardiac output. Clinically, when this happens the patients often complain of a “throbbing” in their neck.

    You can do Lewis leads (if you remember how to do them). On the other hand, I would have tried carotid massage or having the patient valsalva to see if I could get the heart rate slow enough to move the P wave out of the T wave.

  • Glenn Walmsley says:

    Excellent article. I read another viewer give a tip for differentiating rapid AFib from SVT. One tip I use is turning on the monitor volume. It is much easier to hear the irregularity then see it at certain rates. My coworker and I just had what I called an AFlutter this past week, lead II was not diagnostic and a 12 lead was req’d. We did assess for Cannon a waves and I was sure they were present but having never seen them in person I can’t say with 100% certainty. Lastly I have often found slight rate changes with ST when moving a pt/brief breath hold or giving a fluid challenge (should it be required) and helpful in differentiating Reentry/atrial Tachts from ST

    • Great tip! It’s not much of an option prehospital, but one more thing I’ll do when I’m in the ED and a patient has persistent tachycardia is look at the graphic trend of the patient’s heart rate. Reentrant tachycardias will show a stable horizontal line (steady rate), often with a fairly abrupt onset/offset, while sinus tach while drift up and down.

1 Trackback

Leave a Reply

Your email address will not be published. Required fields are marked *