Nitroglycerin - Old and New: Pt 3

I really wasn't planning on doing a part 3, since I had thought there was only so much you could say! A recent discussion on Facebook, however, revealed a surprising amount of enthusiasm for a seemingly boring question:

"SL Nitro tabs or spray?"
It probably matters somewhat whether the NTG is being used to treat ACS or CHF, but that didn't come up in the discussion. Other concerns and opinions, as well as some questions, came up, and I thought it would interesting to see what research was out there to back up some statements or answer questions. 

First off, some straight-forward information of UpToDate about the kinetics of the various forms of nitroglycerin. This shouldn't be viewed as the final answer, but it probably isn't far from it.
 

I'll be using quotations out of the comments from the FB post, since this probably represents the thinking of a good number of people (or at least the thinking of people who like to talk about tab versus spray on a Friday night!

Or a Friday night! (Source)

 Onto the questions!
 
"I heard it works faster? Not sure if its true."
...
 "I have always wondered with the sprays just how much nitro they are getting. I know its supposed to be the same as a tab but I guess I just wonder."
 
Does the spray work faster, or the tablet? Which formulation gives a higher dose? Three studies, all from the 90s, suggest that if there is a difference, it's a small one!

Danish researchers compared 2 sprays of 0.4 mg each versus one 0.5 mg tablet, and checked the blood levels. Not surprisingly, the higher-dose sprays reached a higher dose than the tablets, and faster as well. However, after they corrected for the higher dose, they calculated the time-to-onset and blood levels to be equivalent.

Researchers from Montreal took a different tack. Instead of looking at blood levels, they looked at how quickly nitro spray or tablets could cause arterial vasodilation. Both the tablet and the spray were 0.4 mg. The graph shows the response: note that both the spray and the tablet had each reached their max effect at the 3 minute mark. However, the spray both acted more quickly, and had a more prolonged effect, and also had a higher maximum response.
 

A third group took yet another approach. Instead of looking at brachial artery dilation, or plasma levels, they gave nitro during cardiac catheterizations to directly evaluate cardiac effects. After giving 0.4 mg of the spray or tablet, they found a mixed picture. For example, the spray reduced the LV end-diastolic pressure 30 seconds faster than the tablet, but was 30 seconds slower at lowering the LV end-systolic picture. Basically a wash.

"The tabs have the added problem of the non-english speaking patients not understanding the directions and just swallowing them."  
I would add that it is often difficult to get a English-speaking patient to lift up their tongue. In their defense, how often in your life are you asked to do that? It can seem like an odd, "Simon-says" request. 

Therefore, it's an advantage, I think, that you can just shoot the spray at the tongue, under or over. According to the manufacturer of one brand of nitro spray, it can be sprayed on either area, which ought to make conversations simpler in the back of the rig at 2 AM!

Source

 
"Our pt have dry mouth from being anxious. Tabs don't dissolve in a dry area."
Perhaps the best place to store NTG isn't in a little brown bottle - perhaps it should be kept under the tongues of CHF patients who are struggling to breath. It certainly seems to be the best place to keep a pile of NTG tabs from dissolving!

For example, back in 1986, one doctor found that:

"The addition of 1 ml of saline under the tongue of a patient with visibly dry sublingual tissue will moisten the tissue in preparation for dissolving the nitroglycerin. This simple action has frequently resulted in prompt relief of pain when previous doses of nitroglycerin administered by the patient, and later by hospital staff, had failed."

There also appears to be some science to back up this impression. A group from Japan took an innovative route; instead of looking at critical patients, they had subjects with stable angina ride an exercise bicycle until they felt critical! Well, not critical, but until they felt chest pain. The researchers then did two things. First, they checked how wet or dry the subjects' mouths were. Second, they gave them either nitro tabs SL, or a nitro spray.

It turns out that, for those subjects who had wet mouths, it didn't matter which med they got. But the subjects with dry mouths had their chest pain relieved much more quickly with the spray!

Taking it a step further, another team of researchers looked at giving a teaspoon of water at the same time as giving the SL NTG or the spray NTG. The patients were getting a cardiac cath at the same time (but did not have their mouths checked for wetness!). They found that the patients who got the teaspoon of water with the SL tab had a much greater drop in BP than those who got the SL tab without water. With the spray, however,they got the same drop in BP either with or without the water.

So if you don't have the spray, probably the best idea would be to squirt in some saline or H20 right before you pop in the tabs!

Ask FD to help! (source)

 
"Is paste an option?" ... "Having the paste which allows for slower absorption and removal value where after a spray, there is no taking it back."
Many paramedics have a high regard for the topical ointment preparation of NTG. I'm not sure why, since topical drugs are not very useful in EMS - they absorb slowly, wear off slowly, and have uncertain absorbation, especially when patients are cold, vasoconstricted, or poorly perfusing. The table at the start of the post really illustrates this - it can take 30 minutes to start working, and 7 hours to wear off!

I only found one study that compared intravenous, pill, and paste NTG delivery. They took patients with unstable angina, and split them into 2 groups - IV NTG, or a combination of pills & paste. They adjusted the doses in both groups so that everyone dropped their BP by about the same amount. They found two things.

First, both seemed to relieve the symptoms of angina at the same rate. That's good, because that's the only time I use the paste - when I have a hemodynamically stable patient complaining of mild-moderate chest pain.

Second, they found, unsurprisingly, that IV NTG can achieve a higher blood level than the paste. On the IV stuff, levels of NTG were, on average, around 18 ng/ml, whereas the levels 2 hours after paste application only got up to 1.3 ng/ml.

This issue about the "slow & low" blood levels doesn't matter so much for angina, or even a STEMI. NTG likely doesn't save lives in ACS, and we have other agents that can treat pain. But when EMS is treating a hypertensive CHF patient, they need therapy that works fast, works hard, and that can be "turned off" fast as well. The paste doesn't seem to do any of that.

The Bottom Line
IV nitroglycerin is the ideal EMS drug. 

It works almost instantly, it gets to peak effect almost instantly, it's very good at treating severe hypertensive CHF, and it can be titrated very precisely. Also, we don't need to take the CPAP mask off every 5 minutes to give tablets and water. (You know, they call it continuous positive airway pressure for a reason!)

But until your service can work out the training and supply issues for the IV pumps, and until the cost of the spray comes down a smidge, we may be stuck with the tablets for a while longer. Just understand the differences!

Nitroglycerin - Old and New: Pt 2

A paramedic from AMR in Bridgeport told me about an interesting patient he treated recently. And by "interesting," I mean "briefly terrifying." He was bringing in a 75 y.o. male with chest pain, and had given the patient aspirin, and acquired an ECG. 

I scribbled on it. It's not a clue.

A few things on the ECG bothered him, and the patient's symptoms suggested ischemic symptoms. Since the vital signs were fine, he then gave 1 tab of nitroglycerin (NTG) under the tongue. 

Within 90 seconds, the patient became pale, sweaty, and described feeling pretty awful. The blood pressure bottomed out in the 70s, and the medic noted a drop in the heart rate. He acquired ECG #2:

Small change.

Fortunately, a dram of atropine and a jigger of normal saline rapidly fixed the bradycardia and hypotension. 

Concerned that he had "unmasked" a right-ventricular MI, he then obtained a third ECG, this time with V4R, to interrogate the right side:

TV4 is V4R - rest of precordial leads are the usual.

The rest of the transport was uninteresting, as well as his time in the ED. Although he was admitted, his ECG didn't show any changes from prior, and his troponins were negative. No MI.  
So what happened here? Does the second ECG suggest a cause of the bradycardia? For that matter, what about the first and third ECGs? In what way was the NTG involved?

I'll discuss a few basic questions about NTG-related hypotension, and also discuss an interesting new study out of Montreal, before finishing up with an infrequent adverse effect of nitro.

Why are we concerned about giving NTG during an AMI?
As most paramedic students would be able to tell you, we're worried mostly about MIs that involve the inferior wall, since about half of these involve the RV. Since infarctions of the RV tend to make the patient very sensitive to preload-reducing drugs, it is commonly taught to avoid NTG if there are signs of an inferior infarct. A small study from 1989, for example, looked at a group of patients with diagnosed inferior-wall MI, and found that evidence of RV involvement was strongly associated with NTG-related hypotension.

Many people recommend that you grab an ECG before giving NTG to a patient with suspected cardiac ischemia. While there are different reasons offered for this, the typical reason is to look for an inferior MI, and thus avoid hypotension caused by an RV infarct. 

(On the other hand, Dr Smith is concerned that NTG could "mask" a STEMI - read "Wait until after the ECG to give Nitroglycerin." Pretty dramatic example!)

So, given the apparent importance of this advice, is NTG-related hypotension common?

How often do patients drop their BP after NTG?
Not often!

For instance, a study from 1994 looked at 300 EMS patients that got NTG for CHF or chest pain, and only 4 developed a SBP < 90 mmHg. That's only 1.3%, and even those 4 patients did fine after the NTG wore off.

Another study looked at over 1,500 patients who received NTG from EMS, and only 12 patients had significant adverse hemodynamic effects - only 0.7% of the total.Some folks had some big drops in their BP, but nobody died. Similarly, only 1 patient out of 288 in a third study had hypotension. That patient improved with a 300ml NS bolus.

Okay, so reactions are rare. When these do happen, though, do they predict anything about a presence or location of an MI?

Does a hypotensive reaction to NTG predict an inferior-wall MI? (Breaking news!)
This comes from an abstract presented at the 2013 NAEMSP Scientific Assembly, and so should be considered preliminary. Nonetheless, it appears to be an interesting addition to the nitro literature. (The whole abstract is copied at the bottom of the post)

Researchers in Montreal searched their EMS database for patients who had been transported for suspected ACS. They then picked out the patients who ended up being diagnosed with a STEMI, and who had also received NTG from EMS. Although these patients got prehospital ECGs, the EMTs weren't trained to read them - just to acquire and and transmit. They likely were not able to identify a "likely RV infarct," and were not instructed to withold NTG from anybody based on the ECG. This amounted to about 800 patients over a 2 year period. 

The researchers divide the ECGs into inferior-wall STEMIs and non-inferior-wall STEMIs, and looked at what giving NTG did to either group. They found two things:

• First, patients with an inferior-wall STEMI were more likely to have hypotension initially.  
• Second, after they got NTG, patients with an inferior MI were not more likely to have hypotension, or even a significant drop in BP.

Although you should be cautious if you have a hypotensive patient with huge ST elevations in V4R, it's not clear that the evidence suggests that we routinely need ECGs before giving NTG. 

So what happened to our patient with the bradycardia and hypotension? 

The Bezold-Jarish reflex
He fainted. You can also call it a vasovagal reaction, or a triggering of the Bezold-Jarish reaction, but it's all basically your standard faint

Well, maybe not just like that, but pretty darn close. He never actually "passed out," but given his vitals and symptoms, he wouldn't have stayed upright for long. Fortunately, he was already laying down on a stretcher, so he never lost consciousness. Staying supine when you brady down is a pretty good policy. 

But there's other evidence besides the vital signs that suggest  a vasovagal mechanism, and not a preload-sensitive cardiac ischemia. Take a closer look at the second ECG:

Sinus activity has been totally suppressed, and the rate has dropped down to a typical junctional speed. This is an AV dissociation, rather than 3rd degree block, since the ventricular rate (39) is higher than the atrial rate (0). The QRS hasn't widened at all, suggesting the block is at or above the level of the AV junction. As I noted in a previous post, these are the classic features of a vasovagal bradycardia & syncope.  

The Bezold-Jarish reflex is the likely mechanism, where initial tachycardia triggers a strong burst of activity from the vagal nerve, dropping the blood pressure and suppressing the SA and AV nodes.

From the Cardiology & CCU FB page

The (relative) hypovolemia that you might see when someone gets too much NTG, especially with an RV infarct, would be more likely to produce a compensatory tachycardia, and wouldn't usually produce such dramatic AV blocks.  

Is this something new and rare that I've stumbled across? Sadly, no. There are numerous case reports describing hypotension and bradycardia after giving NTG to patients, many of whom were shown not to be having any MI at all, let alone an RV infarct.

For example, in a 1990 case report, a 36 y.o. male received 2 NTG tabs in the course of a work-up of chest pain. Despite an intially normal ECG, he dropped his BP down to 77/40, and developed a brief brady-asystole. 

His vitals improved without drugs or IV fluids, and his labs and stress test ruled-out MI.

Another case report from 2007 (free access!) describes a similar sudden-onset junctional bradycardia and hypotension in a 60-y.o. male after NTG administration. Further back in ancient history (1981), a case series of 4 patients with this same pattern of bradycardia with hypotension and a narrow-complex bradycardia without sinus activity were described.

In yet another case, a 54 y.o. woman with chest pain (who ultimately ruled-out for MI) was given NTG for chest pain, and developed bradycardia. She described "lightheadedness and malaise," but never dropped her blood pressure.

Atropine (despite the blood pressure) helped the heart rate normalize.

The Bottom Line 
So now that I've described all these nasty bradycardias, should you withhold that next dose? Probably not. Even with EMS giving NTG to thousands of patients in the studies above, the rate of serious adverse effects is ≤ 1%. Those reactions also tend to be transient as well.

I should also point out that the SHCGB Guidelines do not require an ECG before NTG. After all, patients take this medication on their own all the time. On the other hand, they are not calling 911 every day, so the prudent paramedic should be getting ECGs early and often!


________________________________________________________  
103. DOES PREHOSPITAL ADMINISTRATION OF NITROGLYCERIN FOR CHEST PAIN CAUSE HYPOTENSION IN ACUTE INFERIOR WALL STEMI? A RETROSPECTIVE COHORT STUDY
Dave Ross, et al. Urgences-sante; Hopital du Sacre-Coeur de Montreal Montreal
Background. Patients with inferior ST-segment elevation myocardial infarction (STEMI), associated with right ventricular infarction, are potentially at higher risk of developing hypotension when administered nitroglycerin (NTG). However, current basic life support primary care paramedic (PCP) protocols do not differentiate location of STEMI prior to NTG administration.  
Objective. We sought to determine whether NTG administration is more likely to cause hypotension (systolic blood pressure <90 mmHg) in inferior STEMI compared with non-inferior STEMI.  
Methods. We conducted a retrospective chart review of prehospital patients with chest pain of suspected cardiac origin and computer-interpreted prehospital electrocardiograms (ECGs) indicating “acute MI.” Computerized interpretation was performed by the GE Marquette 12SLR-Zoll E Series. Patients were treated by PCPs. We included all local STEMI cases identified as part of a provincial STEMI registry project. Charts were reviewed by trained data extractors using a predefined instruction list. Univariate analysis was used to compare differences in proportions of hypotension after NTG administration, drop in systolic blood pressure greater than or equal to 30 mmHg, and hypotension on initial prehospital blood pressure between patients with inferior wall STEMI and those with STEMI in another region (non-inferior).  
Results. Over a 29-month period, we identified 1,466 STEMI patients. Of those, 798 (54.4%) had complete data and received NTG. Hypotension occurred after NTG in 36 of 461 inferior STEMIs and 29 of 337 non-inferior STEMIs, 7.8% vs. 8.6%, p = 0.69. A drop in systolic blood pressure greater than or equal to 30 mmHg occurred in 23.5% of inferior STEMIs and 23.8% of non-inferior STEMIs, p = 0.91. Initial hypotension was noted in significantly more inferior STEMIs compared with non-inferior STEMIs, 9.9% vs. 4.9%, p = 0.005. Interrater agreement for chart review of the primary outcome was excellent (kappa = 0.94). 
Conclusion. Patients with chest pain and inferior wall STEMI on their computer-interpreted prehospital ECG who receive nitroglycerin do not seem to develop hypotension more frequently than patients with STEMI in other territories, although they are more commonly hypotensive on presentation. Current PCP protocols for NTG administration in computer-interpreted prehospital ECG STEMI appear to be safe.

Nitroglycerin - Old and New: Pt 1

The modern medic gets to use a lot of fancy technology. She interprets capnographic waveforms, reads 12-lead (and 13- and 15-lead) ECGs, places ET tubes, LMAs , and CPAP masks, uses portable lactate analyzers, and even knows how to spell "sphygmomanometer."

Let's face it - compared with all that, nitroglycerin is not a "sexy" EMS tool. 

Unlike this little beauty. You know, I think I've seen this before...

Nitro has been used for angina since 1879, and was also used for "spasms," headaches, and even toothaches. After 133 years of clinical experience, it seems we would have nothing new to say. For example, we take for granted 2 things about NTG:

• The dose is one tab (or spray), every 5 minutes, to a maximum of 3 doses; and
• You should to do an ECG to check for an inferior MI before giving it.

I'll talk about using "high-dose" nitro tablets for CHF in Part 1 today, and discuss the evidence-based concerns about using NTG during an inferior or right-sided MI in Part 2.

Increasing the dose - 1, 2, or 3 tabs sublingual
In case you haven't heard, Lasix is old-fashioned, at least for EMS. It's not a good drug for acute hypertensive congestive heart failure, for any number of reasons, and the modern EMS system uses CPAP and nitro instead.

So how do I usually give NTG for CHF? Once the patient gets to the hospital, I'll use a starting dose of IV NTG at 200 - 400 µg/minute.For the first 20 minutes I'll be adjusting the dose up & down pretty frequently, until the patient's breathing gets better, and the BP has come down.

However, the sublingal NTG dosing used by EMS for CHF is usually the same small dose used for angina pectoris, 0.4 mg every 5 minutes. That's about 80 µg/minute, if you assume that it's absorbed over that 5 minutes. 

Let me point out some basic math:

80 µg < 400 µg

As an emergency doctor, I can get around this problem easily: if I don't have IV NTG available, I give several NTG tabs at once. While the nurses are scrambling to get a second IV and program the IV pump, and the respiratory tech is setting up the CPAP, I'm practically dumping the bottle of NTG under their tongue. The higher their blood pressure is, the more I can give. And I don't stop at 3 doses either; I keep giving it every 5 minutes until the IV juice is flowing.

So, this strategy makes sense pharmacokinetically, and some paramedics having been doing this for years, albeit "off-protocol". For one example, read Peter Canning's excellent blog post NTG and The Hero Medic, about a medic who, evidently, felt comfortable with pretty aggressive NTG doses.

So it's good to read about an EMS system that decided to act rationally, and then study it. In an abstract in the most recent Prehospital Emergency Care, EMS researchers in Buffalo discuss their experience with changing the EMS protocol to allow multiple simultaneous doses of nitro for CHF with hypertension, depending on the BP. I copied the entire article below, but the protocol was simply:

• SBP < 160 mmHg --> 1 tab (0.4 mg)
• SBP > 160 mmHg --> 2 tabs
• SBP > 200 mmHg --> 3 tabs
• Repeat as needed, no maximum # doses

So were patients dropping like flies from this crazy dosing? 

Not so much. Out of 95 patients who got the high-dose NTG, only 3 patients dropped their BP. Which could have been concerning, except that the breathing of those patients was better (which is the whole point), while the hypotension was transient, resolving on the next BP measurement. 

The Bottom Line
Nitroglycerin is an old drug, but we are still learning about how to best use it. Our old habit of "1 q 5', to a max of 3" is perhaps just that, an old habit!

The SHCGB Guidelines, on the other hand, do not set a limit on the maximum number of doses that can be given, and allow the medic to aggressively treat CHF. If the blood pressure is severely elevated during an episode of likely CHF, you can give up to 4 tablets at a time.

Unfortunately, some paramedics may have ingrained patterns of behavior, and there may be doctors who haven't heard about developments in nitro use since 1880. Time to drag them into the 21st century, with all our fancy "new" medications and medical devices!

As for the "sexy" airway tool that looked so familiar above...

Found it!

PREHOSPITAL ADMINISTRATION OF MULTIPLE SIMULTANEOUS NITROGLYCERINE SUBLINGUAL TABS RARELY CAUSES HYPOTENSION
Brian Clemency, Gina Tundo, Jeffrey Thompson, Heather Lindstrom, University at Buffalo, State University of New York
  Background. High-dose intravenous nitroglycerin is a common in-hospital treatment for respiratory distress due to congestive heart failure (CHF) with hypertension. Intravenous nitroglycerin administration is impractical in the prehospital setting. In 2011, a new regional emergency medical services (EMS) protocol was introduced allowing advanced providers to treat CHF with oral nitroglycerin. Patients were treated with two sublingual tabs (0.8 mg) when systolic blood pressure (SBP) was >160 mmHg or three sublingual tabs (1.2 mg) when SBP was >200 mmHg every 5 minutes as needed. To assess the protocol’s safety, we studied the incidence of hypotension following prehospital administration of multiple simultaneous nitroglycerin (MSN) tabs by EMS providers. 
Methods. A retrospective case review of records from a single commercial EMS agency over a six-month period (January–June 2012). Cases with at least one administration of MSN were reviewed. For each administration, the first documented vital signs before and after administration were compared. Administrations were excluded if they were missing pre- or post-administration vital signs. Blood pressure was measured in mmHg.  
 Results. One hundred cases had at least one MSN administration by an advanced provider during the study period. Twenty-five cases were excluded because of incomplete vital signs. Seventy-five cases with 95 individual MSN administrations were included for analysis. There were 65 administrations of two tabs, 29 administrations of three tabs, and one administration of four tabs. The mean change in SBP following MSN was –14.7 (standard deviation 30.7; range +59 to –132). Three administrations had documented systolic hypotension in the post-administration vital signs (97/71, 78/50, and 66/47 mmHg). All three patients were over 65 years old, were administered two tabs, had documented improved respiratory status, and had repeat SBP of at least 100 mmHg. The incidence of hypotension following MSN administration was 3.2%.  
 Conclusion. Hypotension was rare and self-limited in this sample of prehospital patients receiving MSN. 
 

 

"All that wheezes" - CHF and albuterol

Medics aren't happy unless they're arguing about something.

Since most paramedics are 1) intelligent, 2) clever,  and 3) convinced they are more clever than the other medic they're arguing with, they need an appropriate subject to engage in. The ideal topic should provide the opportunity for them to put their knowledge of physiology, gas laws, and hydrodynamics on display, but also allow them to parry with a quick "In my experience..." Ideally, there should be no clear empirical studies on the subject.

The "danger" of giving albuterol to a dyspneic patient who may have CHF is just such a topic.

Fortunately, a complete literature review on the topic easily will fit in one blog post, with plenty of room left over for a relevant case, complete with ECGs!

The Wheezer

"Hey Doc, you gotta see this guy next."

Generally I listen when an experienced nurse tells me that, so I went to see Mr Wheezer right then.

Not our patient.

He did not look good - sweaty, pale, working to breathe, with his second albuterol/ipratropium neb going. "I think it's my allergies," he says. So much for the patient telling me the diagnosis...

So I asked him to tell me about his "allergies." He'd been having some dyspnea on exertion over the past 2 weeks, with a nagging cough. But an hour ago, while at rest, he felt like his chest was being pressed in, right up into his jaw, and he had started sweating buckets. His breathing had worsened, and his wife called 911. He had a history of diabetes, hypertension, and stents in all 3 of the major coronary arteries.

The paramedics had started the series of neb treatments he was receiving, and a quick exam revealed why - he had the loudest, clearest, most unambiguous wheezes I have ever heard. No crackles, no rhonchi, no upper airway crud fouling it up. Textbook wheezes. Even the greenest of EMT-B's, using a Fisher-Price stethescope, could confidently diagnose these wheezes.

"And an S3. He's definitely got an S3."

On the other hand, the first ECG looked like this:

That's pretty bad case of allergies. I looked at his old ECG:

No slam-dunk STEMI here. I grabbed the ultrasound, and took a look at the heart and lungs. No pericardial effusion, no signs of a PE, but when I looked at his lungs, I saw the shimmering artifact that suggests wet lungs. For an example (not my patient):

I took off the neb, and popped in a tab of nitroglycerin. Within 2 minutes he had his color back, he didn't't look so drenched, and he was breathing easier. Like a lightswitch - click! -  he was improving.

I got cardiology involved pronto, before any labs, or even the chest x-ray came back. After they evaluated the patient and the ECG, they shared my concern, and were planning to take him for an urgent cardiac catheterization. Almost as an afterthought, we checked out the labs together.

Troponin - negative.
BNP - negative.

Huh.

We checked out the chest film - clear.

But the ball was rolling, and he went to the lab. Good thing, too, since he ended up having severe stenosis in his mid and distal LAD, as well as the circumflex, which all got new stents. The RCA, which previously been patent, was now totally and permanently occluded.

Soooo, that proves it was CHF, right?!? Well, he still had dyspnea after the procedure, though nowhere near as bad as before, so he got more tests and consults. I won't go through the details, but after being evaluated by 3 cardiologist, 2 pulmonologists, and one lowly ER doc, he had a diagnosis of "likely CHF."

This encounter made me consider a few questions:

How good are paramedics at diagnosing CHF?

The medic in my case only gave albuterol, no nitro or Lasix, and had not obtained an ECG, and so was clearly not considering CHF. But the diagnosis can be tough for physicians, even with all the clinical gizmos at our disposal.

With that in mind, how does the paramedic diagnosis of CHF stack up against the emergency physician's? Turns out, it's pretty good, within limits.

One study from 1995 looked prospectively at the paramedic's diagnosis compared with the ED physician diagnosis. Considering that the doctor had access to medical records, x-rays, labs, etc, the paramedics did fairly well, showing "good concurrence" with physician diagnosis. Another study looked at how well paramedics determined a cardiac cause of dyspnea. This could include angina or MI, as well as CHF, so it wasn't a perfect comparison, but the agreement between medic and doctor had a kappa of 0.71, or "good, approaching excellent."

A more recent, although retrospective, study looked only at the diagnosis of CHF. Interestingly, they studied all the patients whom the paramedic had given furosemide to, using this as a surrogate for a diagnosis. They then looked at the final diagnosis of the the emergency physician. Generally the medics did well, and the doctors agreed in 60 out of 94 cases.

The disagreements are thought-provoking though. Between pneumonia and COPD, a lot of furosemide was given to people to who didn't need it. Furthermore, there is evidence that suggests that treating pneumonia with diuretics is harmful.

Looking towards the future, however, if we can combine capnography and portable BNP analysis in the prehospital realm, EMS could end up being the gold standard for the ED to live up to!


How often does CHF present with wheezing?

Often enough!

In one study of older patients, it was found that about one third of  patients had wheezing with their acute episode. Perhaps not surprisingly, these patients usually were smokers, had a prior diagnosis of COPD, and were using bronchodilators at home. Unfortunately, they did worse with their CHF events, going to the ICU at a higher rate, for instance.

There aren't too many other studies that study the rate of wheezing, but we can also look at the rate of albuterol/beta-agonist administration as a rough equivalent. In one study 20% of patients got albuterol (in addition to other drugs), while in another study only 2% of CHF patients got albuterol (as the sole therapy). 

I think this evidence suggests that there can often be a component of wheezing with acute episodes of CHF, but that "pure" wheezing, without other indications of of CHF, is pretty rare.

Is there a danger in giving albuterol to a patient with CHF?

My patient ended up getting stented for cardiac ischemia. It seems reasonable to wonder if the 2 neb treatments, in addition to being ineffective, might have  exacerbated the ischemia, causing harm. However, looking at the clinical evidence is difficult, as much of it doesn't apply to emergency medicine, let alone paramedicine.

For example, there are a number of studies that analyze the harm associated with use of bronchodilators in patients who have diagnoses of both COPD and CHF. These studies, however, follow patients over months to years, and aren't very relevant.

One ICU study looked at the degree of tachycardia or number of tachyarrhythmias after albuterol neb treatment. They actually did not find much negative effect on the vital signs.

Well, what about studies in the ED or prehospital that look at truly relevant outcomes? Surprisingly, there only appear to be 2 relatively useful clinical studies available to guide us.

The first was a case-control study done in 1992, by Wuerz. They looked at about 500 dyspneic patients who had received prehospital treatments. They found that it was pretty bad to treat asthma or COPD, for example, with Lasix. However, when they looked at the 9 CHF patients who were mistakenly treated with beta-agonists, they found, reassuringly, that "none died."

The second study (Singer 2008) was bigger, using registry data on about 11,000 CHF patients treated in the ED. The nice thing about using such a data source is that you can get a lot of patients, and find associations. The bad news is that you often can't explain what you do find.

Such is the case with this study. About 20% of the patients received beta-agonists, either by EMS or in the ED, but had no pre-existing history of COPD or asthma. They used some statistical rejiggering to try and make fair comparisons, since CHF patients with a history of COPD are not exactly like CHF patients who don't have COPD. I won't bore you with the details, but that's what they mean by "adjusted with propensity analysis" on the following table.

Now, there wasn't any apparent difference in mortality, discharged alive form the ED, or ICU admission - that's what all those big red Xs mean. There was, however, an increased rate of intubation in those CHF patients who were treated with bronchodilators, but who had no history of COPD.Same for BiPAP and inpatient mechanical ventilation.

So what does this all mean? Now, this study wasn't randomized, and it really only shows an association, not cause and effect. The authors state (emphasis is mine):

Inhaled bronchodilator use in these heart failure patients without chronic obstructive pulmonary disease appeared to be associated with worse outcome. Because of the observational nature of these data, we cannot determine whether these patients’ outcomes were worse because they were more severely ill or because of a directly harmful effect of the inhaled bronchodilator.

However, this association persisted after adjustment for propensity score and standard risk factors for mortality. This finding suggests that inhaled bronchodilators may have contributed to the poorer outcomes observed in heart failure patients without chronic obstructive pulmonary disease who were treated with bronchodilators.

Or may not have contributed - we just don't know. The authors allow that treatment with bronchodilators may just a marker for bad CHF, but not a cause of bad CHF. I'm sympathetic to that point of view.

The Bottom Line

 If you're pretty sure the patient has CHF, they need nitro and CPAP. Lasix is old school, and might hurt people when you're wrong about the diagnosis, and there is no evidence that albuterol will help with edema.

On the other hand, if you get fooled by wheezes, don't feel too bad. It might not have helped, but it probably didn't do much of anything at all.