BNP in HF - a smackdown

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

Jeff698

EM/EMS nerd
20+ Year Member
Joined
Aug 12, 2000
Messages
1,998
Reaction score
15
For an interesting editorial smack-down on the use of BNP testing in HF, see the letter from Maisel and the response from Schwam in this month's (June 2005) AEM. Starts on p 572.

Maisel was the lead author of the BNP (breathing not properly...was that a stretch or what) trial and says BNP is a wonderful diagnostic tool for HF.

Schwam disagreed in AEM last year and hence the letter.

It seems to me that Schwam makes a pretty convincing argument. The real entertainment value, however, comes at the end of his reply. He basically accuses Maisel of being a corporation hack.

I'm inclined to agree that moderate BNP values aren't terribly useful as a diagnostic aid in the ED. What do y'all think?

Take care,
Jeff

Members don't see this ad.
 
I have noticed that back when we treated "Acute Decompensated Heart Failure" (which used to be plain old CHF) with Lasix, nitro and oxygen no one really gave it a lot of press. Yet lo and behold that since there is now an expensive new drug available CHF got a new name and I can't go to the mailbox without getting a big, flashy mailer with CDs and CMEs extolling the virtues of the "latest developments" in treating it. It's not about oxygen, I think that's gone generic.
 
I have no connection to Alan Maisel, nor have I read this article in AEM.

That said, I have read most of his papers, and I think that BNP is helpful, although the numbers I'm talking about are as high as 300,000 (seriously). If it's 1000, then I don't know how helpful that is.

THAT said, I had a patient c/o SOB a few days ago. He's s/p heart transplant 7 years ago, and a chest wall lymphoma vs infection (for which he had 5 ribs removed) 2 years ago. His CXR was negative, cardiac markers were negative, and his BNP was negative. The BNP helped me in this case.

Our last journal club on this was July 04, but our experience base at Duke is growing. I am not ready to write BNP off.
 
Members don't see this ad :)
We've discussed BNP a few times before. My own experience is that it is being ordered way more than it is helpful. It seems like every SOB patient is getting one and even patients with known CHF history and physical/workup consistent with a CHF exacerbation are getting one. I think it can be useful when the diagnosis is very unclear but I think most of the time CHF is a clinical diagnosis. I can recall two cases in the last year where it was helpful. The first was an older man with really an excellent history for worsening CHF, decreased exercise tolerance, orthopnea, PND, etc... However, his ekg was normal and cxr was normal except for very minimal cardiomegaly. His BNP was sky high and I think his workup was otherwise pretty normal because he wasn't in decompensated heart failure. In the pre-BNP days he would have been admitted for a rule-out of somewhat atypical anginal symptoms and the echo he almost certainly would have received on hospital day 1 would have shown his idiopathic cardiomyopathy with an EF of 15%. With the BNP we were clued in slightly earlier to the cause of his symptoms but I don't think it really changed anything. Last month I say a young guy with idiopathic pulmonary edema. While running through the list of causes I sent a BNP which was normal and then turned my attention to other causes, toxins, infections, etc... He ended up having HAPE which was high on my list but seemed a bit weird since he had lived at altitude for years and had only gone down to sea level for a few days before returning to altitude.

Unfortunately, I think we are reaching a point where every chest pain/SOB patient gets a troponin,d-dimer, and BNP even if it doesn't make sense. When you throw your net that wide the false positive rate becomes a signifcant problem
 
While we have this test available and have done a journal club, the utility of a BNP in the acute setting seems slim. Certainly there are probably cases like stated where maybe it helped.

Before I am really convinced though, I would like to see some more clinical outcome type numbers. Is it really making that much of a difference in the final outcomes?

The cost is phenomenal and its utility at present seems very limited.
 
roja said:
While we have this test available and have done a journal club, the utility of a BNP in the acute setting seems slim. Certainly there are probably cases like stated where maybe it helped.

Before I am really convinced though, I would like to see some more clinical outcome type numbers. Is it really making that much of a difference in the final outcomes?

The cost is phenomenal and its utility at present seems very limited.


Wasn't there an article published within the last 3-6 months on bnp values and associated out of hospital mortality? I can't remember where I heard about it/read it, but I vaguely remember a cut-off value for bnp and if you were above this value (I think >300) the mortality rate started to increase (measured at one month post d/c)...again, I think...
 
ERMudPhud said:
We've discussed BNP a few times before. My own experience is that it is being ordered way more than it is helpful. It seems like every SOB patient is getting one and even patients with known CHF history and physical/workup consistent with a CHF exacerbation are getting one. I think it can be useful when the diagnosis is very unclear but I think most of the time CHF is a clinical diagnosis. I can recall two cases in the last year where it was helpful. The first was an older man with really an excellent history for worsening CHF, decreased exercise tolerance, orthopnea, PND, etc... However, his ekg was normal and cxr was normal except for very minimal cardiomegaly. His BNP was sky high and I think his workup was otherwise pretty normal because he wasn't in decompensated heart failure. In the pre-BNP days he would have been admitted for a rule-out of somewhat atypical anginal symptoms and the echo he almost certainly would have received on hospital day 1 would have shown his idiopathic cardiomyopathy with an EF of 15%. With the BNP we were clued in slightly earlier to the cause of his symptoms but I don't think it really changed anything. Last month I say a young guy with idiopathic pulmonary edema. While running through the list of causes I sent a BNP which was normal and then turned my attention to other causes, toxins, infections, etc... He ended up having HAPE which was high on my list but seemed a bit weird since he had lived at altitude for years and had only gone down to sea level for a few days before returning to altitude.

Unfortunately, I think we are reaching a point where every chest pain/SOB patient gets a troponin,d-dimer, and BNP even if it doesn't make sense. When you throw your net that wide the false positive rate becomes a signifcant problem
All the cardiologists I know hate troponin. They hate it because it's a test that if positive means an admission, no question. They hate that. They like things better that at noon on an insured pt indicate an admission and work up but at 2 am on a homeless guy are equivocal. If you've been around for a while I'll bet you've had a cardio or IM doc try to get you to send a positive trop home calling it a "troponin leak." I don't fall for that one but with BNP it's tough. What does a high BNP mean if there's no CHF on the x ray and the sat is good? Does a high BNP by itself require intervention? It's like ERMudPhud said "CHF is a clinical diagnosis." What do you do when the marker is positive but the clinical picture does not support it?

BTW This goes to show that casting too wide a net does screw you up. I get stuck with a lot of BNPs and DDimers on people who I wouldn't have gotten them for due to nurses and other docs "protocoling" pts. What do you do with a positive DDimer in a pt with low prob? If you ignore it and they coincitentally have a PE, DVT or develop one in the future you're screwed.
 
I was just in a morning report session with a cardiologist who was talking about CHF and BNP, how to treat. He suggested that those who are "warm and wet" with an elevated BNP, should be left in the ER for 23h put on nesiritide and then d/c'd later. He feels that these pts are unnecessary admissions. it doesn't seem very practical to me to let someone diurese in the ER. I am just wondering if this is being implimented by anyone?
 
In places with Observation units that would be a typical admission - mild CHF that you diurese and send home. It's unreasonable to expect the ED to board them in the ED proper though, we have patients to see.
 
Apollyon said:
I have no connection to Alan Maisel, nor have I read this article in AEM.

That said, I have read most of his papers, and I think that BNP is helpful, although the numbers I'm talking about are as high as 300,000 (seriously). If it's 1000, then I don't know how helpful that is.

THAT said, I had a patient c/o SOB a few days ago. He's s/p heart transplant 7 years ago, and a chest wall lymphoma vs infection (for which he had 5 ribs removed) 2 years ago. His CXR was negative, cardiac markers were negative, and his BNP was negative. The BNP helped me in this case.

Our last journal club on this was July 04, but our experience base at Duke is growing. I am not ready to write BNP off.
I hope you're not trying to say that you didn't clinically diagnose this CHF episode in a patient with a BNP of 300,000.

I think BNP is useful in COPD'ers v. CHF'ers who have a vague presentation. Otherwise, I think it's an expensive test that only confirms (and often delays) a clinical diagnosis.
 
docB said:
BTW This goes to show that casting too wide a net does screw you up. I get stuck with a lot of BNPs and DDimers on people who I wouldn't have gotten them for due to nurses and other docs "protocoling" pts. What do you do with a positive DDimer in a pt with low prob? If you ignore it and they coincitentally have a PE, DVT or develop one in the future you're screwed.

In our system, only a physician can order a d-dimer. When ordering it, the physician must go through a computerized version of the Wells criteria. If it scores 3 points or more, it forces you to perform definitive testing and will not allow you to order the d-dimer. Of course there are workarounds to this -- you can always send the blood down and then call the lab 20 minutes later to add it on, but most people take a hint when the computer system won't let them order it.

On the topic of CHF, do any of you use bedside echos and other forms of ultrasound to assess CHF? For those iffy cases where I'm not sure, I've found that taking a look at the IVC and doing a bedside echo helps.

If you hear bibasilar crackles in a patient with a known history of CHF, but who presents with cough and fever, taking a quick look at the IVC and seeing it collapse with every breath gives more evidence of the patient having pneumonia instead of CHF. Of course, this doesn't eliminate CHF entirely as the patient may have isolated left heart failure, but it does make me less likely to rush to administration of furosemide.
 
southerndoc said:
I hope you're not trying to say that you didn't clinically diagnose this CHF episode in a patient with a BNP of 300,000.

No - my point was some people can be mildly or moderately SOB with a fever, and can have a BNP that is off the charts. Is there really a difference between a BNP of, say, 15000 or 20000 and a level of 150000? How much more fluid can there be in the lungs? Likewise, I've seen slam-dunk CHF's with BNP of <2000. It's a tool, but not nearly the whole boat.
 
kbrown said:
I was just in a morning report session with a cardiologist who was talking about CHF and BNP, how to treat. He suggested that those who are "warm and wet" with an elevated BNP, should be left in the ER for 23h put on nesiritide and then d/c'd later. He feels that these pts are unnecessary admissions. it doesn't seem very practical to me to let someone diurese in the ER. I am just wondering if this is being implimented by anyone?
Based on this does he still feel that way?

JAMA. 2005 Apr 20;293(15):1900-5.

Short-term risk of death after treatment with nesiritide for decompensated heart failure: a pooled analysis of randomized controlled trials.

Sackner-Bernstein JD, Kowalski M, Fox M, Aaronson K.
 
MS05' said:
Wasn't there an article published within the last 3-6 months on bnp values and associated out of hospital mortality? I can't remember where I heard about it/read it, but I vaguely remember a cut-off value for bnp and if you were above this value (I think >300) the mortality rate started to increase (measured at one month post d/c)...again, I think...


Well, an EF of 15% is pretty much a kiss of death in terms of mortality. What I meant by clinical outcome numbers was, does the use of BNP significantly alter clinical outcome from the ED?

:)
 
roja said:
Well, an EF of 15% is pretty much a kiss of death in terms of mortality. What I meant by clinical outcome numbers was, does the use of BNP significantly alter clinical outcome from the ED?

:)

Its sort of frustrating that in the era of, "evidence based medicine," it is impossible to separate data from ego. When I sat (not too long ago) in the morning report room as a third year medical student, I endured listening to cardiologist after cardiologist bash EM physicians over the improper ordering and administration of BNP. Depending upon an individual clinician's specialty and scope of practice, BNP was viewed simultaneously as, "costly," "useless," "wonderful," and "of questionable value."

Because it is a relatively new modality receiving a great deal of press, its probably equally important to take a step back from the literature and relate the use of BNP to human physiology. From what I understand (and please correct me when I go too far off of the beaten path), BNP is an endogenous hormone that counterregulates the catecholamine and RAA cascade. Natural opposition to catecholaminergic cascade is beneficial in CHF because the patient is already in extremis. Continued hypoxemia and increasing myocardial oxygen demand further stimulates the release of catecholamines and can potentiate cardiovascular collapse and dysrhythmia. The administration of loop diuretics often activates the RAA via volume depletion and can have unintended effects. Nesiritide's intrinsic natriuretic properties work even in patients with less than optimal renal function and do not cause compensatory activation of the RAA axis. Proponents of BNP argue that cost effectivness of their drug is actualized through the potential decrease in hospital length of stay and admission. Medicare patients afflicted with CHF are over age 65 and are very likely to return to the ED within 30 days. Theoretically, adequate treatment/diuresis with nesiritide lowers the "CHF recidivism" rate and therefore reduces costs. I understand that medicare will only pay for ONE CHF exacerbation within a 30 day period.

From my interaction with attendings and sub-par understanding of the literature, it appears that BNP values have a role in the exclusion of CHF. BNP is quite sensitive above the upper limits of normal (admittedly a controversial discussion) and can assist the emergency clinician with excluding CHF in the elderly patient presenting with dyspnea. Whatever your kidney function is, a BNP value of several thousand usually indicates systolic dysfunction. Similarly, a BNP less than 100 can help to move CHF farther down the differential list. Obtaining a BNP just because someone presents with SOB is probably not the best decision. What does one do with a "moderately" elevated value found in a patient with a past medical significant for MI and CRF ?

This discussion is just plain cool because it highlights just how much the emergency physician can, "fix." I recall my paramedic days when aggressive diuresis and preload reduction could make someone "turn the corner" and stave off an intubation. Recent studies now support the use of aggressive nitroglycerin dosing and diuresis. When I encountered my first patients on the medicine wards, I was surprised at the complete lack of consensus and understanding with regard to patients in CHF. Doses of tridil infusions varied immensely, as did the dosing of PO and IV lasix. Just as the heart failure discussion has moved in the direction of expensive endogenous hormones, it has recently doubled-back into the terrain of good old oxygen. While the debate over administration of nesiritide rages on, what about the use of CPAP as a temporizing measure prior to NTI or OTI ? CPAP is inexpensive, easy to administer, and has the potential to decrease BOTH ICU admission AND endotracheal intubation.

Though this BNP learning curve is rather steep, I hope most emergency clinicians can agree that patients who are in CHF and volume overloaded will benefit from:

-aggressive nitroglycerin dosing
-adequate doses of diuretics
-decrease of myocardial oxygen demand and reduction of RAA/catecholamine cascade

Whether this is achieved through more conventional treatments or via newer modalities like nesiritide infusion is better left to residents and attendings with much more experience in the realm of CHF than myself. In the meantime, I'll continue enjoying the discussion and watching my free CD's from W. Frank Peacock, MD and the Cleveland Clinic. :)
 
I typically find BNP doesn't help ME, it helps the admitting team who seems to need it in order to function these days. You know... hypoxia, rales, DOE, PND, pulmonary congestion on CXR aren't enough to make the diagnosis - but if the BNP is elevated...

I have started using BNP to track my chronic CHF'ers for level of decompensation. We have a few people that come in every few days because they don't feel comfortable at home or don't understand their disease. Now, its pretty tough to send home someone with a CAD/CHF history c/o SOB. But, if the exam doesn't support it, the CXR hasn't changed, the enzymes are negative and the BNP is down, I feel a bit better.

Anyone else tracking it in these chronic patients?
 
What about using troponin as a prognostic indicator in heart failure? Do you all think it has any value?
 
Scrubbs said:
I typically find BNP doesn't help ME, it helps the admitting team who seems to need it in order to function these days. You know... hypoxia, rales, DOE, PND, pulmonary congestion on CXR aren't enough to make the diagnosis - but if the BNP is elevated...

I have started using BNP to track my chronic CHF'ers for level of decompensation. We have a few people that come in every few days because they don't feel comfortable at home or don't understand their disease. Now, its pretty tough to send home someone with a CAD/CHF history c/o SOB. But, if the exam doesn't support it, the CXR hasn't changed, the enzymes are negative and the BNP is down, I feel a bit better.

Anyone else tracking it in these chronic patients?

In the old days, like up to a few years ago, we made this decision based on physical exam, 02 requirements, EKG, and CXR. No evidence of ACUTE CHF and no increase in 02 requirements you go back home. Any evidence that addition of BNP to this decreases the rate of inapropriate ED discharges?
 
Apollyon said:
Some of our patients have more than 50 recorded
:eek:

I hope Duke is getting a volume discount from whomever makes their BNP assay.
 
Top