The use of non-invasive positive pressure ventilation (NIPPV), especially CPAP, in decompensating patients with acute cardiogenic pulmonary edema (ACPE), can reduce the need for intubation, resolve symptoms faster, and reduce the hospital/ICU length of stay and mortality significantly. Only 13 patients with ACPE need to be treated in order to save one extra life, and 9 patients to avoid an extra intubation.

The Setup:
EMS has just unloaded a patient with a long-standing history of coronary artery disease, prior MI/bypass surgeries and congestive heart failure. He has been feeling progressively short of breath for a few days, with decreasing exertional tolerance and increasing orthopnea and paroxysmal nocturnal dyspnea. He denies any chest pain, fever, or cough. His vitals are as follows: heart rate 105 regular, respiratory rate 24, blood pressure 185/95, oxygen saturation 84% on room air, and he is afebrile. Your chest examination reveals coarse crackles to the apices of both lungs, and pitting, lower extremity edema extending to both knees.

While initiating your diagnostic investigations, you begin aggressive medical treatment of the patient’s presumed congestive heart failure/pulmonary edema. He seems noticeably fatigued, however, and you are concerned that he may be sliding into respiratory failure. He may soon need to be intubated, but your ICU currently has no capacity for intubated patients. A colleague asks whether your patient would be a good candidate for NIPPV.


The ACEP Clinical Policy for ED patients with ACPE and NIPPV has received a Level B recommendation to “use 5 to 10 mm Hg CPAP by nasal or face mask as therapy for dyspneic patients with acute heart failure syndrome without hypotension or the need for emergent intubation to improve heart rate, respiratory rate, blood pressure, and reduce the need for intubation, and possibly reduce inhospital mortality” Silvers 2007. How does NIPPV improve respiratory status in CHF? ACPE results from LV dysfunction leading to increased back pressure on the pulmonary circulation and extravasation of fluid into the alveoli. This fluid may overwhelm the ability of pulmonary lymphatics to offload, subsequently diluting surfactant and decreasing lung compliance and alveolar ventilation from collapse. This increases the patient’s work of breathing considerably. In addition, upright positioning of the ACPE patient causes gravitational pooling of fluid into the bases, which may increase ventilation-perfusion mismatching and shunting. Resulting hypoxia may trigger anxiety and sympathetic discharge, resulting in tachycardia, hypertension, peripheral vasoconstriction and diaphoresis.

NIPPV alleviates many of these problems by maintaining positive airway pressure and preventing end-expiratory alveolar collapse, increasing lung compliance and decreasing work of breathing, while decreasing cardiac preload and afterload. NIPPV can also drive fluid out of the alveolar spaces back into the vascular/lymphatic tree, benefiting alveolar ventilation and reducing V/Q (ventilation/perfusion) mismatch/shunting Bradley 1991, Baratz 1992. For the patient, NIPPV reduces the work of breathing, hypoxia and anxiety.

The Evidence:
Vital FMR, Saconato H, Ladeira MT, Sen A, Hawkes CA, Soares B, Burns KEA, Atallah ÁN. Non-invasive positive pressure ventilation (CPAP or bilevel NPPV) for cardiogenic pulmonary edema. Cochrane Database of Systematic Reviews 2008, Issue 3. Art. No.: CD005351. DOI: 10.1002/14651858.CD005351.pub2.

This review has compiled all previous randomized controlled trials on the use of NIPPV in treating acute cardiogenic pulmonary edema (ACPE). NIPPV has been previously shown to be of benefit in reducing intubation rates, in-hospital mortality and hospital length of stay in acute exacerbations of COPD. In the current review, the authors examined potential benefits of NIPPV in treating ACPE for similar benefits. As with all Cochrane reviews, there was no industry funding or bias concerns raised. A funnel plot analysis showed no evidence of publication bias, and some heterogeneity initially noted was resolved when 1 quasi-randomized trial was excluded from analysis.

The authors identified 21 trials with 1071 adult patients enrolled. Most of the included trials were conducted in the ED, with some reporting ongoing care into ICU and other settings. The included trials used comparisons of NIPPV with standard medical treatment to medical treatment alone. Standard medical treatment included the use of: diuretics (furosemide), morphine derivatives, nitrates, vasopressors (dopamine, epinephrine, norepinephrine), inotropes (dobutamine, digitalis) and other agents, used variably at the discretion of treating physicians. NIPPV decreased intensive care unit length-of-stay by 1-day (95%CI -1.60 to -0.53 days), but did not effect hospital length-of-stay. The Table summarizes the impact of NIPPV on other important outcomes.

CPAP and BIPAP were not equivalent. Whereas CPAP significantly reduced mortality in undifferentiated ACPE patients (RR 0.56; 95% CI 0.35 to 0.89), BIPAP did not (8 trials, 373 patients, RR 0.70; 95% CI 0.40 to 1.23). Similarly, CPAP reduced intubation rates (RR 0.46; 95% CI 0.32 to 0.65), but BIPAP did not (8 trials, 373 patients, RR 0.68; 95% CI 0.27 to 1.73). However, ACPE patients with hypercapnia have reduced mortality and intubation rates with BIPAP than those treated with CPAP.

The authors concluded that NIPPV, and especially CPAP, in addition to standard medical treatment, is effective in reducing patient intubations, ICU length of stay and mortality rates, without any increased risk of precipitating AMI. The low number-needed-to-treat (NNT) suggests that this is a high yield intervention in the appropriate candidates who may otherwise need to be intubated.

Outcome                Relative Risk (95% CI)       
Mortality                0.60 (0.45-0.84)
Intubation              0.53 (0.34-0.83)
Acute MI Rate         No difference

Some limitations for this review were acknowledged. There were multiple small trials included, with small numbers of outcome events, thereby limiting the inferences made about these outcomes. Considerable variation existed in medical treatments used, which may have biased some of the results.

Candidates for NIPPV
The patients included in these studies were adults with evidence of acute or acute-on-chronic pulmonary edema, using criteria defined by the AHA and the European Society of Cardiology, and these were the criteria used for defining ACPE in this Cochrane review. These included clinical features of ACPE (increasing dyspnea, cough with or without sputum production, pallor, cyanosis, normal or elevated blood pressure, rales, cold clammy skin, and other signs of impending respiratory failure). These clinical findings were supplemented by appropriate findings on chest X-ray, electrocardiogram, serum markers for AMI or CHF or echocardiography.

NIPPV would be contraindicated in any situation of excessive facial injury/bleeding/secretions, hypovolemic hypotension, potential airway/lung injury and failing to maintain mentation or airway protection. Additionally, delaying endotracheal intubation for the use of NIPPV in patients too ill to benefit has been associated with patient deterioration.

A validation of prior research
A number of high quality reviews/meta-analyses on this topic have generated similar conclusions for reducing mortality, intubation rates and hospital and/or ICU length of stay, particularly in support of CPAP for ACPE Collins 2006, Masip 2005, Peter 2006, Winch 2006. On the contrary, the recently published Three Interventions in Cardiogenic Pulmonary Oedema (3CPO) trial (NEJM 2008; 359: 142-151) was a large multicentre trial with 1069 randomized patients to NIPPV with standard medical care vs. standard medical care alone which was not included in the Cochrane review. The C3PO trial showed no difference in mortality or intubation reduction, but faster resolution of respiratory symptoms, abnormal vital signs, and metabolic disturbances (hypercapnia, acidosis) with NIPPV. C3PO may have been limited, though, by the exclusion of severely ill patients, the failure of almost 20% of enrolled patients to complete their assigned treatments and cross-contamination between the treatment arms. Additionally, the protocol used no objective criteria for intubation, and the reported intubation rate (3%) is much lower than those in other large trials/reviews (~12%).

The practice environments for many of these studies varied from Emergency Departments to ICU settings, which may alter the outcomes of the data based on selection bias, illness severity (spectrum bias) and how aggressively they are treated (co-intervention bias). The current Cochrane review, however, had a number of studies conducted in the ED, where the reported benefits appear to be robust. The key for NIPPV in ACPE is administration early in the patient’s clinical course. Unfortunately, there were no studies included in the Cochrane review that examined the role of NIPPV in the prehospital setting Kosowsky 2001.

Case Outcome:
You quickly decide that an early trial of NIPPV is warranted in your decompensating patient with ACPE. A trial of CPAP is initiated using 5-15cc H2O, and the patient’s breathing starts to become easier as he concurrently begins to vigorously diurese. His vital signs stabilize, and his metabolic derangements start to resolve. He is subsequently admitted to a step-down area in your hospital for imminent withdrawal of NIPPV and ongoing care. One less ICU bed blocked!

Dr. Upadhye is a co-chair of BEEM (Best Evidence in Emergency Medicine) and an assistant clinical professor of emergency medicine at McMaster University.



# William A. Hardman, M.D. 2010-01-21 03:12
Good article, but he makes the mistake that was disproven about 20 years ago- CPAP does not force fluid from the alveolus into the circulation. It has its effect by changing preload,afterlo ad, and improving LV function, thus facilitating fluid resorbtion by Starlings forces.

Add comment

Security code

Popular Authors

  • Greg Henry
  • Rick Bukata
  • Mark Plaster
  • Kevin Klauer
  • Jesse Pines
  • David Newman
  • Rich Levitan
  • Ghazala Sharieff
  • Nicholas Genes
  • Jeannette Wolfe
  • William Sullivan
  • Michael Silverman

Earn CME Credit