Better late than never comes April’s offering!
Perhaps one of the most formidable opponents we have to face in critical care is that of ARDS and as such any intervention that we can utilise to combat it shouldn’t be ignored. But how do we know which ones to use once we’ve turned down the tidal volumes? Oscillation? Nitric? ECMO? Prone positioning?
I thought we’d spend a bit of time on this subject over the next few months revisiting some of the evidence as our approach to this condition in many ways captures what’s best about our speciality; namely an ability to turn physiological principles into evidence based treatment strategies.
So have a read of this NEJM paper on prone positioning (http://www.nejm.org/doi/full/10.1056/NEJMoa1214103)
Alternatively check out this excellent summary from Dr Rebecca Lathey and see what you think…..
“The paper describes a multicenter RCT taking place in 27 ICUs mainly in France (one in Spain) involving 446 patients, all of whom were diagnosed with severe ARDS and were intubated and ventilated for this for a duration of less than 36 hours. Of these patients, they were randomized into two groups as per a computerized protocol; one of which continued ‘supine’ (in fact, semi recumbent) positioning, whereas the other were pronated for at least 16 consecutive hours within an hour of randomization. Between the groups, efforts were made to standardize other variables of care; similar ventilation settings etc. From this point, the protocol regarding pronation becomes slightly more complicated, in that there were multiple criteria for stopping pronation, and that patients in the supine group could be pronated if clinically indicated. A huge amount of data from patients in both groups was recorded at variable stages which included physiological criteria, admission data (to allow adjustment for variables), use of other techniques, complications and various radiological/biochemical tests. Data was analysed using a variety of statistical tools and compared to the pre-defined outcome measures: mortality, extubation rate, length of ITU stay and complications amongst others. Results were described as:
* Reduced rates of ‘rescue therapies’ such as ECMO, nitric oxide and almitrine bismesylate use in pronated patients. However, there were slightly higher use of NMBs and a longer sedation time.
* Pronated patients had better oxygenation (higher Pa02: Fi02 ratio) with a lower Fi02 and PEEP required at days 3 and 5
* 28 day mortality was nearly halved in the prone group versus supine, and this reduced mortality ‘persisted’ at day 90.
* There was a higher rate of successful extubation in the prone group (80% vs 65%)
* A similar complication rate, length of stay, use of NIV post-extubation and tracheostomy use between both groups.
From initial reading, this sounds like a clear suggestion that proning patients with severe ARDS early post-intubation should be adopted into daily practice. However, the article also raises a few points that suggest its results may not be easily applicable to all (including our) ICUs…
1. All of the ICUs enrolled in the study were experienced in proning patients ( ie. it is ‘used in daily practice for more than 5 years’). One would imagine that the complication rate etc would be higher in those units not so experienced in the technique.
2. The article is not specific about the duration of proning that is required to create the above results. It also used some subjective criteria including clinician discretion about when to abandon proning. If a patient’s proning was abandoned during the trial, were they then moved into the supine group for results analysis?
3. Whilst the results are quite specific in describing better oxygenation at days 3 and 5, it does not fully describe the effects on any other than these days!
4. Whilst randomization was eventually computer-based, exclusion of patients was presumably manual. This included over half of the patients with ARDS during the time frame not being screened. It could be suggested that the decision of who to screen or later who to exclude was open to selection bias.
5. The patients in the supine group appeared to have more severe organ failure (SOFA score), and required a higher proportion of vasopressors which could impact on mortality difference.
6. Finally, I would be interested to see the difference in outcome when comparing pronation with ECMO, as this appears to be more widely adopted in patients with severe lung pathology on our unit. In summary, this is an interesting article, which does provide evidence that proning specific patients (severe ARDS) at an early stage could reduce mortality. However, the logistics of this and generalizability to other units is questionable. Significantly, other trials have shown no difference in mortality, and there are certainly confounding variables in this trial to explain why they found such a difference. Therefore, pronating patients could emerge as common practice in such patients (although, how often do we see patients meeting the severe ARDS criteria that is diagnosed within 36 hours to make this the case?) but ONLY if future studies support these findings on a large and generalizable scale.”