Can eliciting Homan's sign actually CAUSE a PE?

[QuantumX]

An EM resident told me a few weeks ago that trying to elicit Homan's sign on a patient with suspected DVT can actually cause a thrombus to dislodge and cause a PE.

Can anyone comment on that? I couldn't verify it from any textbook..

---

Members don't see this ad.

 

[Purifyer]

An EM resident told me a few weeks ago that trying to elicit Homan's sign on a patient with suspected DVT can actually cause a thrombus to dislodge and cause a PE.

Can anyone comment on that? I couldn't verify it from any textbook..

Definately _can_ according to what I've been told. It's in my textbook too..

 

[Sessamoid]

Definately _can_ according to what I've been told. It's in my textbook too..

On top of that, it's a useless sign. Poor sensitivity and specificity. Even the physician who found it and for whom the sign is named has disavowed the validity of it.

 

[Adawaal]

On top of that, it's a useless sign. Poor sensitivity and specificity. Even the physician who found it and for whom the sign is named has disavowed the validity of it.

Yep. A good history and a coin flip is about as informative....

 

[kinetic]

Homan's sign is Murphy's Law.

If you don't do it because data has shown it to be both nonspecific and insensitive, you'll get pelted with fecal matter for skipping "a basic and easily performed part of the physical exam in a patient with complaints that are suspicious for DVT."

If you do it because you've been trained like a chimp to check for it, you'll get lectured on "the insensitivity and nonspecificity of this test, which many students and residents can't seem to understand."

You're better off stabbing your eyes out with a rusty pick.

 

[DrQuinn]

Agree with kinetic 100%. As a student you should do it. When you present it, I can guarantee a surgery/IM/EM resident/attending will stop you and question you about it or comment on it. Be sure to note that it is horrible insensitive/unspecific.

It could dislodge a PE.

Remember that recent studies are showing that early ambulation actually shows a decreased incidence of PEs (contrary to nursing/PT/OT belief).

Q, DO

 

[MustafaMond]

I disagree with kinetic.

You should not do "Homan's sign" because it is an extremely $hitty test.
We never encourage students to do it at my program.

If you want to get kudos from the attending, just fast-track a venous doppler ultrasound and get the answer that way.

 

[QuantumX]

Thanks guys. I think I'll take Mustafa's advice and NOT do it anytime..
🙂

 

[Old MD]

All those who said that Homans can cause a PE, can you provide some evidence? Purifyer what book were you referring to
Thanks.

 

[Old MD]

Mustafam do you order a doppler study on all your suspected PEs? What is the sensitivity?

 

[MustafaMond]

Mustafam do you order a doppler study on all your suspected PEs? What is the sensitivity?

I was talking about using doppler to r/o DVT, not PE.
And I order them on any and all suspected DVTs.

Sensitivity of compression doppler for DVT is VERY high (positive predictive value of 94%)

A discussion on "What is the best test for PE" is another discussion, that would be best argued on a separate thread. (My opinion being Spiral CT)

 

[kinetic]

A discussion on "What is the best test for PE" is another discussion, that would be best argued on a separate thread. (My opinion being Spiral CT)

Gold standard is still pulmonary angiogram. Spiral CT is easier to perform, but not available at all centers and can still be misinterpreted.

 

[DrQuinn]

Gold standard is still pulmonary angiogram. Spiral CT is easier to perform, but not available at all centers and can still be misinterpreted.

There is no spoon!

Q, DO

 

[Old MD]

I was talking about using doppler to r/o DVT, not PE.

agreed. asked because your first post didn't specify.

Anyone with evidence that Homan's is harmful?

 

[pikachu]

I just looked on Pubmed and couldn't find any case reports of a PE actually being precipitated by someone testing the Homan's sign. If you google it, though, there are some medical websites that claim you can precipitate a PE by doing it. http://www.gpnotebook.co.uk/cache/-1831862264.htm
Not sure if this assertion is grounded in any actual cases, or if it just seems like it could be true.

The American College of Phlebology says, "DVT is often first noticed as a progressive "pulling sensation" in the posterior aspect of the lower leg, although the onset of pain may be acute and excruciating. Symptoms may worsen with ambulation. Calf tenderness and a positive "Homan's sign" (pain with dorsiflexion of the foot) are absolutely unreliable in making the diagnosis of DVT."

Also, it's interesting to note that the venous Doppler scan is much less sensitive in asymptomatic patients (sensitivity is 93% if pt is symptomatic but only 59% in asymptomatic patients).

 

[Seaglass]

positive predictive value of 94%

I'm no Dr. Statistics Ph.D., but wouldn't a test with high sensitivity have a high negative predictive value. If you want a high positive predictive value then you want a test with good specificity.

C

 

[Tenesma]

actually while pulm angio is the gold standard it is usually the test of last resort ....

 

[Sessamoid]

Also, it's interesting to note that the venous Doppler scan is much less sensitive in asymptomatic patients (sensitivity is 93% if pt is symptomatic but only 59% in asymptomatic patients).

Not surprising at all, if you think about. The population of patients who have symptomatic DVTs are more likely to have significant obstruction of flow than those who don't. Bigger obstructions and bigger clots == easier to visualize on doppler scan.

 

[Sessamoid]

I'm no Dr. Statistics Ph.D., but wouldn't a test with high sensitivity have a high negative predictive value. If you want a high positive predictive value then you want a test with good specificity.

C

A high sensitivity doesn''t mean a high negative predictive value. High negative preditive value only answers the question, "If I have a negative test result, what are the chances that the pathology is truly absent?" You can easily have a test that is highly sensitive, but with low negative predictive value, depending on the prevalence of the disease. Takes this example:

You have a disease with 80% prevalence, so you have 100 people of which 80 have the disease. You also have a test which is 95% sensitive. which only tells you enough about the test to know that 76 of the diseased people will test positive, and 4 will test negative. What sensitivity doesn't tell you is how many false positives there will be. Supppose that the same test also has 40% false positives, meaning that of the 20 who don't have the disease 8 of them will test positive, and 12 will test negative. Of the 16 patients that test negative, only 12/16 or 75% will truly be free of disease. A decent number, but not fantastically high. In this case,a high false positivity rate combined with a high prevalence can decrease the overall number of negative results to the point where the number of false negative test results actually make up a significant proportion of all negative test results.

So while high sensitivity tests tend to have high negative predictive values, that may not always be the case, depending on your subject population and false positive rate of the test.

 

[Seaglass]

Yes, I am aware of how prevalence affects NP value, I was making the clarification that tests with high sensitivities generally do not have high PPV as a characteristic.

C

A high sensitivity doesn''t mean a high negative predictive value. High negative preditive value only answers the question, "If I have a negative test result, what are the chances that the pathology is truly absent?" You can easily have a test that is highly sensitive, but with low negative predictive value, depending on the prevalence of the disease. Takes this example:

I agree completely, but I was comparing specificity to sensitivity with regard to PPV - ie. in the same population the test with the highest specificty has the best PPV versus the test with the best sensitivity (e.g. D-dimer vs. CT)

 

[Old MD]

It could dislodge a PE. (sic)

Q, DO

Definately _can_ according to what I've been told. It's in my textbook too..

On top of that...

To all those who claimed that doing the Homans is dangerous because it can dislodge a thrombus, what is your evidence?

thanks

 

[MustafaMond]

To all those who claimed that doing the Homans is dangerous because it can dislodge a thrombus, what is your evidence?

thanks

Mustafam do you order a doppler study on all your suspected PEs? What is the sensitivity?

The thread is about HOMAN's SIGN!!
Now, I don't know what med school you went to, but where I work, they don;t use doppler to Dx PE.
Are you asking whether we routinely doppler the LEs of pt's with PE? You don't have to, but yes. I've seen it done a lot. It gives you a baseline, so that when they have been on coumadin for 2 months, and then you do a doppler, you will know for sure that there is treatment failure (indication for umbrella)

The data is all over pubmed about how $hitty Homan's sign is.
Whether or not it causes PE, who cares...its a craptastic test that is a waste of time.
Its a waste of time, and possibly dangerous (according to many books and websites)
Look up the data yourself, if you have such a problem with that.

 

[Tenesma]

Doppler for PE??? what? who? that is just plain silliness... here is one of many references: Turkstra F, Kuijer PM, van Beek EJ, Brandjes DP, ten Cate JW, Buller HR. Diagnostic utility of ultrasonography of leg veins in patients suspected of having pulmonary embolism. Ann Intern Med 1997;126:775-81.

That paper basically shows there is NO diagnostic utility of doing u/s for PE.

In fact there are some studies that show up to 30% of proven PEs (by angio/spiral CT) have absolutely normal Proximal Dopplers...

A good read is the PIOPED study (early 90s) as background - but a lot of movement has occured away from V/Q towards the better Spirals.

There is NO evidence regarding homan's sign/massage/serial compression devices dislodging clots that lead to PEs - primarily as it would be difficult to do a well informed consent. IE: patient with warm, red, edematous leg s/p surgery - "Sir, we want to enroll you in a study where we massage your calf for a determined amount of time to see if we can dislodge a lot"... in fact, in the literature there isn't even anecdotal evidence.

 

[Old MD]

Mustafa: you need to calm down and quit postin w/out thinking. You wouldn?t have to remove too many feet from your mouth that way.

My question was directed to those who said that doing the Homans is dangerous because it can cause the thrombus to be dislodged. This was the ORIGINAL POSTER?s question. Two people specifically answered yes, it can (Quinn and Purifier). Sessamoid made good points about other aspects of the test, but he too agreed that doing the Homans can dislodge the clot.

My question to them was very simple: what is the evidence that it can.

I asked because when I was a resident I looked this up. As far as I can see there IS NO EVIDENCE ANYWHERE IN THE LITERATURE that this is true (Medline, Sneiderman?s bibliography, most modern physical diagnosis textbooks). Of ANY kind (case reports, case series, trials)

Tenesma (what a name!) has made the same point in a good post.

Regardless, trying to get a Homans sign in a patient suspected of either DVT or PE is not advocated. That is because it has poor sensitivity (actually not necessarily a bad thing for a clinical maneuvre) AND poor specificity (meaning even if you elicit the sign, it could be any number of things).

My point in asking these individuals the specific question is to point out that in medicine YOU NEED TO BE CLEAR IN YOUR HEAD ABOUT ***WHY*** YOU DO THINGS (or don?t). You do not skip a Homans because ?it is dangerous? ? thats an old wives tale. You skip it because it is little value in telling you what you want to know. This example (homans sign) is a small issue, but having good habits in medicine is a good thing.

I never asked you any question about this in the whole thread ? the only question I asked YOU was whether you do a leg Doppler in all your patients with PE (a stupid thing to do, IMO). You made allusion to doing dopplers in your first post, but you werent clear about the circumstance, so I asked you politely, and when you explained I agreed with what you said. Didn?t you read that before you posted?

To Purifier: though you didn?t answer I think I know what book you are talking about. You are from New Zealand, so it is a safe bet you are talking about Talley and O?Connor?s textbook. I have only seen it recently, and it is mostly excellent. Unfortunately not too popular with students here (when I was in school we all used Bates, and it is still the standard). Talley however is the ONLY place I have seen in print this claim that Homans dislodges clots. It is an unreferenced statement. If you look at the primary literature, it is difficult to find any evidence (of course I suppose one could say it has never been formally reported).

Sessamoid, you must be using Sapira (or have been taught by someone using it, maybe). Sapira talks about Dr. Homan (a Harvard surgeon who trained under Cushing and worked at Brigham) not reffering to his own sign eponymously. You will note Sapira does not say that doing the manoeuvre is dangerous.

 

[Sessamoid]

I'm aware that there is no literature to support the potential hazard of performing a check for Homan's sign. My initial post was not clear enough in that regard. "On top of that", should have been "In addition to the possible concern about dislodging a clot".

I'm not citing Sapira, just oral teaching probably a few degrees separated from the primary source (i.e. Dr. Homan). In other words, it's hearsay from my side.

 

[Pilot Doc]

A friend of a friend of a friend story:

2nd year med student is sent to perform his first "complete" H&P on an inpatient. He seeks to elicit Homan's sign (which is positive) and the patient promptly keels over dead from a saddle embolus. The unfortunate student recounts this story as an attending years later while teaching physical diagnosis to assure his green students that their first H&P will go better than his did.