Introduction with a case
Some years ago I admitted a 50-year-old otherwise healthy woman with hypoxemic respiratory failure from H1N1 influenza. Initially she was doing well with high-flow nasal cannula, on 60% FiO2 and 40 liters/minute flow. However, over the course of several hours she became progressively hypoxemic, requiring 85% FiO2 to maintain a saturation of ~90%. Despite worsening hypoxemia, she looked remarkably well: she was in no respiratory distress, mentating normally, with no complaints.
Arguably I should have intubated her, because she had profound and worsening hypoxemia. However, intubation would have led to worsening hypoxemia, paralysis, and a long ICU course. She just looked too good – I couldn’t bring myself to intubate her. Instead, I did something a bit unusual – I asked her to try lying on her belly. She rolled into a prone position with minimal assistance, and her saturation jumped to 98%. Within hours her FiO2 was weaned down to 50%.
Over the next few days we encouraged her to spend time in a prone position. She made an uneventful recovery, with a total ICU length of stay of about 5 days.
Scaravilli et al. 2015: Prone positioning improves oxygenation in spontaneously breathing non-intubated patients with hypoxemic acute respiratory failure: A retrospective study.
This is a retrospective case series describing 15 non-intubated patients with hypoxemic respiratory failure who underwent proning. In general, most patients had pneumonia (13/15), were immunocompromised (8/15), never required intubation (13/15), and survived (12/15). Proning was performed 43 times, a median of twice per patient. Oxygen was most often provided with a facemark, BiPAP mask, or helmet CPAP. Proning was performed for a median of 3 hours and a maximum of 8 hours.
During eighteen proning procedures, the patient was maintained on the same amount of respiratory support throughout the entire procedure. Proning improved oxygenation transiently, but this decreased to baseline afterwards (adjacent figure). Proning had no effect on PaCO2, pH, respiratory rate, or hemodynamics. Two prone procedures were interrupted due to patient intolerance, but otherwise no complications were noted.
This study is limited by its retrospective design. In particular, it is unclear what selection criteria were used to determine which patients underwent proning. However, measuring serial timepoints within each patient’s course allows every patient to serve as their own internal control.
The universality of prone physiology
Typically we prone intubated patients. However, the physiology of proning should work regardless of intubation status. This is based largely on:
- Improved secretion clearance (gravity works in your favor)
- Recruitment of posterior lung regions which often become atelectatic
- Improved ventilation / perfusion matching
Previous studies have found that proning intubated patients improves oxygenation, although there is often a decline in oxygenation after return to the supine position. Scaravilli 2015 found the same pattern when proning non-intubated patients, suggesting that the same physiology is at work.
Among intubated patients, mortality benefits were not found until proning protocols were developed incorporating a longer duration (“dose”) of proning, typically 18 hours per day. If this were also true among non-intubated patients, then meaningful benefits might only occur among patients who could tolerate proning for prolonged periods.
However, even if non-intubated patients couldn’t tolerate 18 hours of proning per day, a few hours of proning could be sufficient to recruit the lung bases. This may be analogous to a recruitment maneuver in an intubated patient. In theory, this could be followed with other strategies to keep the lungs inflated (e.g. BiPAP).
Risk vs. benefit?
The main risk of awake proning could be that it delays intubation until the patient has deteriorated further and is more profoundly hypoxemic. Such a delayed intubation could increase the risk of peri-intubation desaturation. In order to avoid this, close monitoring could be used with prompt intubation if the patient were continuing to decline.
Ultimately the benefits of proning depend on the patient’s overall trajectory. In some patients (such as the introductory case), a brief period of proning may avert the need for intubation. However, other patients have a more progressive decline, inevitably requiring intubation. In these latter patients, delaying intubation is pointless and dangerous. Careful attention to the patient’s trajectory and underlying illness may help clarify the ideal treatment.
Which patients might be candidates for awake proning?
Proning an awake patient may be used only in carefully selected patients with intensive monitoring. This could be considered in the following situations:
(1) Isolated hypoxemic respiratory failure without substantial dyspnea (the “paradoxically well appearing” hypoxemic patient). A reasonable candidate might meet the following criteria:
- not in multi-organ failure
- expectation that patient has a fairly reversible lung injury and may avoid intubation
- no hypercapnia or substantial dyspnea
- normal mental status, able to communicate distress
- no anticipation of difficult airway
(2) Patients who do not wish to be intubated (DNI). The main risk of awake proning is that it could cause excessive delays in intubation. In the DNI patient who is failing other modes of ventilation, there is little to be lost by trialing awake proning (1).
(3) This could be attempted as a stop-gap measure for a hypoxemic patient when intubation isn’t immediately available (e.g. desaturation during transportation). Many awake patients are capable of proning themselves, so this could be achievable without any resources.
Why don’t we prone awake patients more often?
From the pulmonary standpoint, supine positioning may be the worst possible position. Supine positioning may promote aspiration, as gravity tends to pull oral secretions towards the larynx. Supine positioning promotes atelectasis of the posterior lung segments (which are larger and more important than the anterior segments). Among obese patients, abdominal contents compress the diaphragm when supine, further promoting atelectasis. Finally, expectoration is difficult in a supine position, as the patient must expel secretions against gravity. Dr. Levitan refers to the supine position for intubation as the “coffin position.”
This isn’t anything particularly new. The dangers of remaining in a supine position were recognized over a century ago, when the term “hypostatic pneumonia” was used to describe bedridden elderly who succumbed to dependent pneumonia (Mackenzie 2001). Alternatively, the value of occasionally proning hospitalized patients has long been recognized by respiratory therapists (as a component of postural drainage therapy). This may deserve further consideration today: even with modern intensive care, we can’t escape our evolutionary past.
- Prone positioning has been shown to improve oxygenation and survival among intubated patients with ARDS.
- Scaravilli et al. 2015 proves that proning awake patients will similarly improve their oxygenation, although this improvement is only temporary.
- Proning awake patients may occasionally be a useful technique to recruit the lung bases, improve oxygenation, and promote secretion clearance.
- Among patients with hypoxemic respiratory failure, it remains unclear which patients could be treated with noninvasive techniques (e.g. high-flow nasal cannula and awake proning) versus which patients should be intubated. If noninvasive techniques are attempted, this should be done with intensive monitoring and the ability to intubate promptly if necessary.
- Of course, careful consideration must be given to whether the patient is comfortable, and whether or not ongoing therapy is overly burdensome.
Image credits: Massage image from wikipedia here.
Latest posts by Josh Farkas (see all)
- PulmCrit- Epinephrine vs. atropine for bradycardic periarrest - February 13, 2017
- PulmCrit- Six myths promoted by the new surviving sepsis guidelines - January 30, 2017
- PulmCrit- How to convert a VBG into an ABG - January 16, 2017