Why CPAP works in acute heart failure and other bedtime stories

The potassium issues from the previous post resolved with IV magnesium, and the patient was discharged on oral magnesium indefinitely. It is ironic that oral magnesium supplements cause diarrhoea, which is the most common source of magnesium loss.

I have been wandering around the hospital looking for cases to learn from. It’s almost like being an safari, knowing that there’s probably something fascinating out there if you look long and hard enough. Last Friday there was an ITU admission of dermatomyositis with respiratory failure, which brought up plenty of discussion about NIV and invasive ventilation from the medical students. Let’s go through what we found out.

A 65 year old man presented with worsening fatigue and weakness for the previous 10 days, with a rash occurring at the same time.

The weakness seemed to affect “every movement”, and he was brought in by ambulance when his partner felt he looked awful and was poorly responsive.

By the time I saw him, he had already been on high dose steroids, and the most spectacular aspects to his rash were apparently gone. Some scaling from resolving Gottron papules was still visible around the knuckles, but there was no rash around the eyelids and his upper chest and face were normal.

This was not the classical presentation. The textbooks will speak of a proximal weakness that spares the distal muscles. Our patient seemed to have a particularly rapid progressing variant.

Anti Jo-1, which is present in only 20-30% of cases of myositis, signifies a poor prognosis. Our patient was negative for this. A CT abdo/thorax also looked for any suspicious lesions causing dermatomyositis as a paraneoplastic syndrome and was clear.

His CK was elevated around 5000, and was coming down. In myositis, the CK levels can be used to assess the response to therapy, as the muscle inflammation is of a background, continuous sort, leaking to a steady leak of CK that will reduce as the myositis is treated. In contrast, in rhabdomylosis following say a fall, the CK rise occurs as a response to the initial injury, which is usually one off. In rhabdomyolysis, CK rises within 12 hours of the muscle injury, peaks at 1–3 days and then declines 3–5 days after the original muscle insult. The peak CK level in rhabdomyolysis can be used to guide treatment and assess prognosis, but daily CK levels are less helpful for assessing the response to treatment once it has been shown to be falling.

I had never known myositis to be so serious as to cause respiratory muscle failure, but apparently respiratory compromise is not uncommon. Patient UK does not mention respiratory complications from myositis. I found a short paper on three case reports where patients with myositis needed ventilation support, which really helped me appreciate high-end spectrum of this disease.

This got me thinking more broadly about the indications for invasive ventilation.  I was wondering why this patient was not a candidate for NIV instead, given he was making efforts and there was no airway obstruction and there was no vomiting/secretions into the airway. Perhaps it was because of his depressed consciousness, leading to a potentially poor gag and risk of aspiration? I’ll have to find out from the anaesthetists on Monday.

I found this handy summary of the indications for intubation, which basically splits it into:

  1. Obtaining and maintaining an airway
  2. Correcting gas exchange
  3. Protecting from aspiration.

NIV can correct gas exchange, but cannot maintain an airway nor protect from aspiration.

The medical students also asked why CPAP is used for acute heart failure leading to pulmonary edema. In all honesty, I realised that I did not understand this well enough to teach it with confidence. I had a simple understanding that CPAP literally blows the fluid back into the pulmonary vasculature from the alveoli/interstitium.  I decided to read it over the weekend and explain it on Monday. I’d like to write down it while I understand it, and would love to hear if anyone else has heard any other explanations.

There are two problems:

  1. Fluid is in the alveoli and in the interstitium of the lungs. This is impairing gas exchange.
  2. The left ventricle has a limited amount of contracting power, so the cardiac output is not enough to get fluid out of the lungs. Ultimately, to fix the pulmonary edema we need to get the left ventricle’s cardiac output back up.

So what determines the cardiac output in a failing heart?

Well, we know that cardiac output is stroke volume x heart rate. This means we need to increase the stroke volume, the heart rate, or a combination.

The problem with increasing the heart rate is that this will impair myocyte perfusion, as increases in the heart rate decrease the proportion of the time the heart spends in diastole (when the coronary arteries perfuse the myocardium) whilst increasing the oxygen demand. The rate can only go so far.

On an aside, it’s worth pointing out that the diastolic pressure, which seems to be treated as the curious sidekick to the heroic systolic blood pressure, actually determines cardiac perfusion. In a tachycardic septic patient, you will never get systolic blood pressure up if the coronary perfusion pressure (diastolic pressure minus right atrial pressure) is 15mmHg.

So what else determines the stroke volume, and hence cardiac output?

When the preload is less than what the left ventricle is good enough to deal with, preload determines cardiac output.

This is the case in healthy hearts at most preloads. The left ventricle in Mo Farah can deal with pretty crazy amounts of preload thrown at it, and will just pump it straight out. No pulmonary edema for him.

When the preload is greater than what the left ventricle can deal with, afterload determines the cardiac output.

This is what has happened to our heart failure patient. The preload determines the amount of myocyte stretch you have just before the myocytes contract. In a heart failure patient, there simply isn’t the intrinsic contractility to deal with stretch. The myocytes will follow the normal Starling laws upto a point (say  15mmHg preload). After that, any further increase in the preload (say to 18mmHg) will only produce the contractility that their maximum contractility (at 15mmHg) will allow. You therefore have a limited cardiac output as preload increases, which means fluid congests in the inflow to the left atrium.

So, we need to find a way of increasing cardiac output. We can’t increase the preload. What we can do instead is reduce the afterload. When the afterload is reduced, the left ventricle can eject blood more rapidly, and empty a greater proportion of its volume into the aorta. This increases its stroke volume. It’s like squeezing a full water balloon connected to a hose full of high pressure water versus squeezing a full water balloon connected to an empty hose. For the same effort, you will empty much more of the water balloon into the empty hose.

This means our only ward-based option is to reduce afterload. We could go to ITU and use ionotropes to increase the contractility, but this comes with risks (especially arrhythmias), increases the myocardial oxygen demand and may worsen mortality.

CPAP has three key effects in acute left ventricular failure:

1. The extra intra-alveolar pressure counteracts the extra hydrostatic pressure in the pulmonary vasculature that was causing fluid to translocate into the alveoli and interstitium.

2. The increase in thoracic pressure gives a boost to the contraction of the LV ventricle. You could argue that it also impairs filling, but the effect on boosting contraction in a failing heart with no shortage of filling pressure is more significant.

3. The thoracic pressure is increased but there is no pressure increase outside the thorax. This is the equivalent from the LV’s point of view of pressure around the thorax staying the same, but the pressure outside the thorax being lowered. This will feel like a reduced afterload outside the thorax to the LV.

For these three reasons, CPAP is the king when it comes to pulmonary edema from heart failure.