Mechanical ventilation for pulmonary oedema

Mechanical ventilation for pulmonary oedema has never come up in the CICM Fellowship Exam, perhaps because it is something clearly indicated and well supported by the evidence.

In summary:

  • High flow nasal prongs
    • Do not improve clinically significant outcomes
    • May be an option in cases where CPAP (NIV) is contraindicated
    • Probably better than conventional (unhumidified) high-flow oxygen
  • NIV:
    • Decreases mortality
    • Decreases intubation rate
  • Intubation and invasive ventilation:
    • Often not required
    • Has the added advantage of decreasing the distribution of cardiac output to the respiratory muscles 

High flow nasal oxygen for pulmonary oedema

In spite of the fact that they supposedly produce a small amount of PEEP, high flow nasal prong oxygen is not normally indicated in pulmonary oedema. Inata et al (2017) found that in the paediatric cardiac surgery setting the HFNP does physiological things one might expect from PEEP, such as increasing FRC and improving cardiac output. However, this has not been demonstrated in adults, let alone adults in acute pulmonary oedema. Chua & Kuan (2017) considered whether HFNP were “ready for prime time” in the treatment of acute decompensated heart failure, and concluded that “maybe” was most likely the answer.

In terms of empiric outcomes, we have nothing supportive. Makdee et al (2017) trialled the use of HFNP in acute cardiac pulmonary oedema and found that it decreases the patients’ respiratory rate when it is compared to patients who are on normally delivered oxygen. The primary outcome was respiratory rate at 60 minutes following the commencement of therapy, and this indeed improved by a statistically significant 3.3 breaths per minute, from 25.1 down to 21.5. This is another example of a statistically significant trial finding having minimal clinical relevance.

Overall, it is not clear that the evidence supports this technique, except where the decision is between HFNP and conventional wall oxygen. One is aware of the dangers of using dry cold wall gas, and one can predict that wherever oxygen is indicated, it would be better to have it warmed and humidified. However, if the situation would conventionally call for NIV, then putting the patient on high flow nasal prongs is not supported by any studies. It is difficult to draw conclusions from the Makdee trial, as those patients were insufficiently ill to merit NIV (out of the 128 patients with pulmonary oedema only four ended up requiring NIV or intubation).

Non-invasive ventilation

In short, this whole section can be summarised as “it works”. A meta-analysis by  Peter et al (2006) and the later systematic review by Vital et al (2013) concluded that NIV has significant benefits:

  • NIV clearly improves mortality
  • NIV clearly reduced the need for intubation

CPAP or BiPAP? This is a frequently asked question. In short, it does not seem to matter, nor should there be much difference if you think about it logically. Positive pressure is the most important aspect, and the effect is closely related to the mean airway pressure, which in turn is closely related to PEEP. Ergo, adding some extra pressure support during the inspiratory phase may have some additional pressure-related beneficial effects, but really all the heavy lifting is done by the PEEP. A PEEP-heavy mode of ventilation like CPAP is also the best-studied modality, whereas bi-level support has not had as much attention from investigators.

Another frequently asked question is, why does everybody pick 10 cm H2O as the PEEP level to start the CPAP? Mathru et al (1982) and Kontoyannis et al (1991) seem to be the earliest papers regarding this. Both investigated patients with poor systolic function and both concluded that a PEEP of around 10 was an optimal balance.

So, if NIV is so good for APO, why might one need to resort to invasive ventilation? There are several predictors to describe patients who are failing NIV. Ferreira et al (2013) and Masip et al (2013)  found that patients who failed had the following characteristics:

  • Hypercapnia
  • Lower pH (7.25 and below, even among normocapneic patients)
  • More severe hypoxia (P/F ratio under 190 or so)
  • Acute myocardial infarction (vs. more chronic heart failure)
  • An LV ejection fraction below 30%
  • Systolic blood pressure lower than 140 mmHg

In addition to needing to be rescued after failing a trial of NIV, there are other advantages to having invasive mechanical ventilation in pulmonary oedema.

Invasive mechanical ventilation for pulmonary oedema

The addition of invasiveness to mechanical ventilation has some added benefits, on top of what positive pressure alone can accomplish. Most of these advantages are actually unrelated to the pressure. The main benefit is sedating and paralysing the patient for intubation, which has the effect of decreasing the total body oxygen demand. In addition, the poorly compliant lungs are no longer being ventilated by the patient's own effort, and this decreases cardiac demand further. This widely believed factoid is based on a dog study by Viires et al (1983), who produced cardiac tamponade in the test animals and then demonstrated that the resulting shortness of breath created a redistribution of around 21% of cardiac output to their respiratory muscles. 

The severity of respiratory failure does not determine survival; the severity of the poor cardiac function does. Fedullo et al (1991) found that non-surviving patients had both pulmonary oedema and hypotension. In addition to decreasing preload by positive pressure, improving afterload has an additive effect. Adding an IABP was found to be beneficial by  Kontoyannis et al (1999). That was the only study of invasive mechanical ventilation for cardiogenic pulmonary oedema which looked at survival (it improved, at least to discharge).


Makdee, Onlak, et al. "High-flow nasal cannula versus conventional oxygen therapy in emergency department patients with cardiogenic pulmonary edema: A randomized controlled trial." Annals of emergency medicine 70.4 (2017): 465-472.

Chua, Mui Teng, and Win Sen Kuan. "The use of high-flow nasal cannula in acute decompensated heart failure: ready for prime time yet?." Journal of Emergency and Critical Care Medicine 1.9 (2017).

Inata, Yu, and Muneyuki Takeuchi. "Complex effects of high-flow nasal cannula therapy on hemodynamics in the pediatric patient after cardiac surgery." Journal of intensive care 5.1 (2017): 30.

Peter, John Victor, et al. "Effect of non-invasive positive pressure ventilation (NIPPV) on mortality in patients with acute cardiogenic pulmonary oedema: a meta-analysis." The Lancet367.9517 (2006): 1155-1163.

Vital, Flávia MR, Magdaline T. Ladeira, and Álvaro N. Atallah. "Non‐invasive positive pressure ventilation (CPAP or bilevel NPPV) for cardiogenic pulmonary oedema." Cochrane Database of Systematic Reviews 5 (2013).

Masip, Josep, et al. "Risk factors for intubation as a guide for noninvasive ventilation in patients with severe acute cardiogenic pulmonary edema." Intensive care medicine 29.11 (2003): 1921-1928.

Fedullo, Anthony J., et al. "Acute cardiogenic pulmonary edema treated with mechanical ventilation: factors determining in-hospital mortality." Chest 99.5 (1991): 1220-1226. 

Kontoyannis, D. A., et al. "Mechanical ventilation in conjunction with the intra-aortic balloon pump improves the outcome of patients in profound cardiogenic shock." Intensive care medicine 25.8 (1999): 835-838.

Mathru, Mali, et al. "Hemodynamic response to changes in ventilatory patterns in patients with normal and poor left ventricular reserve." Critical care medicine 10.7 (1982): 423-426.

Wiesen, Jonathan, et al. "State of the evidence: mechanical ventilation with PEEP in patients with cardiogenic shock." Heart 99.24 (2013): 1812-1817.

Ferreira, Catarina, et al. "Predictors of failure of noninvasive ventilation in acute cardiogenic pulmonary edema." (2013): P2487.