Non-invasive Ventilation in Severe Community Acquired Pneumonia

An elderly man attended the Emergency Department with a 3-day history of increasing breathlessness, a cough productive of white sputum and intermittent left anterior chest heaviness. He had no significant comorbidities, and lived independently. On examination, Sats were 80% on high flow face mask O2, RR was 20 and there were bilateral crepitations to both mid-zones. ABG showed type one respiratory failure with pO2 5.7 and Lactate 2.0. CXR revealed left-sided consolidation and bilateral upper lobe venous diversion with small bibasal pleural effusions. HR was 75, BP 102/57 with an elevated JVP. Temperature was 38 with WCC 14.6. An ECG showed sinus rhythm with inferior T wave inversion. Urea was 20 and Creatinine 105 (baseline 59). Troponin was 2.9 (normal range <0.05). The impression was of left-sided community-acquired pneumonia and acute renal impairment with Congestive Cardiac Failure (CCF) likely precipitated by a recent myocardial infarction. He was admitted to ICU, given broad-spectrum antibiotics and treatment for acute coronary syndrome and he was commenced on Non-Invasive Ventilation (NIV). He required FiO2 0.8 and PEEP 10 to maintain pO2 >8. Echocardiography showed at least moderate LV systolic impairment, moderate MR and Grade 3 diastolic dysfunction. He was commenced on Noradrenaline and Dobutamine infusions. On day 2, he became profoundly bradycardic and had an asystolic cardiac arrest. Spontaneous circulation was restored following 1 minute of CPR and Adrenaline 1mg IV. He was intubated during the arrest. Despite ROSC, he continued to deteriorate with increasing pressor requirements and worsening AKI . After discussion with family, treatment was withdrawn.

What is the evidence for using non-invasive ventilation in patients with severe community acquired pneumonia?

Craig Walker

Non-invasive ventilation (NIV) describes the use of assisted mechanical ventilation via techniques not requiring endotracheal intubation. Strictly, it denotes inspiratory support above a background PEEP but CPAP is usually counted together with Bi-level support as NIV. Utilisation varies greatly between geographical areas and between hospitals, but one study found that its use increased from 4% of patients on ventilators in 2001 to 11% in 2004 ⁽¹⁾. It is being increasingly applied in settings other than intensive care units such as recovery rooms, cardiology, neurology and oncology wards; even in some palliative care units ⁽²⁾. Its main applications are for exacerbations of COPD and cardiogenic pulmonary oedema (when Continuous Positive Airway Pressure (CPAP), rather than true NIV, is often used). The evidence for reducing the need for intubation and improving survival is the most robust for exacerbations of COPD (as has been demonstrated in systematic reviews)⁽³⁾ but some have advocated its use for other forms of acute respiratory failure, such as pneumonias, asthma and acute respiratory distress syndrome.

Community-acquired pneumonia (CAP) remains a huge cause of morbidity and mortality worldwide. In the United States, for example, it leads to more than 500,000 hospital admissions per year. Globally, it remains the leading cause of death from infection in the developed world ⁽⁴⁾.

Antonelli and colleagues conducted the only randomised trial investigating NIV as an alternative to endotracheal intubation in patients with severe hypoxaemic respiratory failure who failed to improve despite medical therapy ⁽⁵⁾. In the single-centre Rome study published in 1998, they recruited 64 patients. There was a non-significant improvement in PaO2:FiO2 ratio in both groups over the first hour (p=0.21), 10 of 32 patients in the NIV group subsequently required intubation and there was no significant survival difference between groups. More patients in the intubated group had serious complications (66% vs. 38%; p=0.02) and had pneumonia or sinusitis related to the ETT (31% vs. 3%, p=0.003). In those who survived, the NIV group had shorter periods of ventilation (p=0.006) and shorter ICU stays (p=0.002).

Confalonieri and colleagues randomised patients with severe CAP admitted to 3 Italian ICUs to receive either NIV or standard medical therapy with supplemental oxygen ⁽⁶⁾. 28 patients were in each group and, although they found that those randomised to NIV had lower rates of intubation (21% vs. 50%, p=0.03) and shorter ICU stays (p=0.04), subgroup analysis revealed that only those patients with underlying COPD benefited; there was no improvement or trend toward improvement in the non-COPD patients.

Small observational studies have suggested that, in the absence of COPD, NIV is not likely to be useful in preventing intubation for hypoxaemic respiratory failure, with high intubation rates (16 of 24 patients in one study), although the authors stated that it may be worth a trial period in closely-monitored settings ⁽⁷⁾.

Hilbert and colleagues performed a randomised trial of NIV in patients immunocompromised due to haematological malignancies, transplantation or HIV presenting with fever, pulmonary infiltrates and acute respiratory failure. There were 26 patients in each arm and those who received NIV had significantly lower intubation, serious complication and mortality rates than those given standard therapy (8).

Lessons Learned

In appropriate patient groups (such as those with exacerbations of COPD), NIV can improve survival and decrease hospital stay. However, its use in severe community-acquired pneumonia cannot be routinely recommended given its high likelihood of failure and a lack of evidence to demonstrate improved outcomes. If used in the ICU setting, patients who do not subsequently require intubation may have improved outcomes in terms of duration of stay and hospital-acquired infections. However, the high likelihood that endotracheal intubation will be required should be borne in mind and preparations should be made accordingly.

References