Prone Ventilation

The physiology is as follows: when a patient is moved right and left through an arc, injurious cytokines are distributed diffusely throughout the lungs. The blood flow can shift around as well. Kinetics distribute the injury into smaller, more manageable units across a larger surface area. Additively, prone positioning applies gravity and further unloads the compressive weight of anterior fat, cardia, and abdomen from the lungs. Further, the prone position, combined with an unrestrained abdomen, pulls the anterior diaphragm away from the mediastinum, effectively increasing negative pleural pressure and promoting an open lung expansion aided by gravity.

Anatomically, a human being’s airways are designed to drain from posterior to anterior. Leaning forward actually promotes gravitationally assisted pulmonary toilet, but moving into a supine position puts the body at a disadvantage. Although the consequences are not nearly as severe for humans, some animals, such as turtles, will suffocate if left on their backs. If they cannot right themselves, they cannot clear their own secretions.

Although the medical community often touts mortality as the gold standard, that is inconsistent with patients’ goals. Quality of life is what more than 99% of patients cited as their goal in the Study to Understand Prognoses and Preferences for Outcomes and Risk of Treatments, otherwise known as SUPPORT.² Additionally, mortality data are lacking in a large body of ICU interventions. Among them are pulmonary artery catheters, pulse oximetry and targeting oxygen (O₂) saturation >92%, continuous renal replacement therapy, postpyloric feeding, and physical therapy. None of these has ever proven to save lives. The point is, clinicians do many things that are not mortality-guided, and yet they are done because they are the right thing to do. I propose that the application of kinetic/prone ventilation is another modality that lacks mortality data but improves our patient care.

Before delving into the details of KPT – formerly referred to as continuous lateral rotation therapy (CLRT) – clinicians may benefit to know a brief history of it. In the 1940s, after World War II, turning the patient over every 2 hours became a standard nursing practice, particularly out in the field. The intent was to limit bedsores. In the 1970s, the practice was touted as being beneficial to pulmonary toilet, and the first commercially available automatic-turn bed was marketed. The next 2 decades saw a plethora of underpowered studies suggesting the benefit of KPT/CLRT.

So where did the practice of ventilating patients in the supine position come from? The answer is elusive, but I suspect the issue boils down to the simple fact that it is more convenient for nurses to assess their patients when the patient is supine. Somewhere along the way, supine ventilation became accepted as the right thing to do. Also, it is perceived to be somehow cheaper. However, none of this makes sense. There is no literature that supports or hypothesizes that the static supine position correlates with patient benefit.

On the other hand, a meta-analysis dating back to 1992 showed that KPT is beneficial. ³ The 6 studies involving a total of 419 patients indicated a 50% reduction in hospital-acquired pneumonia, a 35% decrease in ventilator days, and a 24% reduction in ICU length of stay (LOS) for patients who were treated with KPT. To put the ICU-LOS reduction with KPT in some context, it is greater than that cited by the Leapfrog Group⁴ for the use of intensivists, which reduces ICU LOS by 9% to 18%. The authors of the 1992 meta-analysis called for a large randomized, prospective trial to be done, and many followed.

One randomized, controlled trial⁵ demonstrated a 42% reduction of ventilator-associated pneumonia (VAP) when patients received KPT instead of the standard of care (eg, turning patients every 2 hours). In a prospective case-control study,⁶ investigators turned patients 30°, not quite ìkineticî by the Center for Disease Control’s (CDC) definition, and saw a 35% reduction in ventilator-associated pneumonia (VAP). But, like the 1992 meta-analysis of Choi and Nelson,³ Kirschenbaum and coworkers⁶ did not note a significant effect on mortality.

Data suggest that even if only the head of the bed was elevated, it would result in a 22% reduction of VAP.⁷ However, compliance with this maneuver across the United States is poor. Interestingly, Ahrens and colleagues⁵ noted that compliance with the standard of care for turning patients every 2 hours was poor as well.

To be fair, no gain in mortality from KPT has been proven to date. In fact, one review showed that the pooled analysis for mortality data from the various studies might even lean slightly toward the side of harm.⁸ I will place an astrix on this pooled analysis, in that nearly every study has been applying KPT to advanced respiratory failure and more often than not, utilizing KPT as a ìrescue therapy.î The same review showed that the total VAP reduction, when pooled as a whole, is 52%. Also, hospital costs are reduced by between 5% and 24%. Additionally, the Choi and Nelson data,³ which is already 15-plus years beyond its intitial publication, demonstrated a 24% reduction in ICU LOS without meeting the Leapfrog criteria.⁴

KPT is an easy maneuver that can be readily applied to any ICU, and in an era of accreditation and public reporting through organizations such as the Joint Commission and of even the federal government looking at VAP rates, it is an important maneuver to know how to do. At the expense of a very small mortality, there is a significant impact on VAP, which directly affects patients’ quality of life, LOS, rehabilitation potential, and further long-term disability.

Prone Versus Lateral Turning

So is prone ventilation superior to lateral turning or Kinetic therapy?

Even before double-blinded, randomized controlled trials were done, the benefits of prone positioning were known intuitively, and that knowledge was not limited to clinicians. As an example, the first thing runners do after crossing the finish line is to bend forward and put their hands on their knees to try to get better air. Adult patients with chronic obstructive pulmonary disease and pediatric patients with asthma alike get into a forward-leaning tripod position when faced with respiratory distress. Without thinking, everybody knows this; it is just a matter of proving it with science.

In 1974, the first prone proposal was published in the field of pediatrics. In a commentary, the author hypothesized that prone ventilation makes sense and that clinicians should conduct the scientific trials to prove it.⁹ It was not until the late 1990s, however, that there was a second flurry of interest. Along came 5 major, fairly large studies with similar protocols examining prone ventilation. Some of them looked at KPT as well. A meta-analysis of these studies reveals various trends with regard to prone positioning (Table 1).

Table 1. Head-to-Head Summary

For example, the studies with the best results get the patients prone sooner than later, similar to treating pneumonias with appropriate antibiotics administered early. Do not wait to apply prone positioning as a rescue therapy because it actually may be better applied early on.

Another trend shows itself in the duration of prone ventilation. The longer patients are in the prone phase, the more improved mortality rates. For example, Mancebo and colleagues¹⁰ place severely ill patients, with Simplified Acute Physiology II scores >50, on their face for 20 hours a day.¹⁰ That study was the only one that demonstrated a statistically significant mortality benefit. Patients ventilated prone had a 43% ICU mortality rate, compared with 58% for patients ventilated supine.

Overall, the longer the patient was prone, the better the outcomes in all parameters across the board, with the exception of skin breakdown. On the basis of the findings of these studies, it seems that it may be beneficial to quickly turn patients prone, especially the severely ill ones. For other patients, at the very least laterally turn them, even if it is not prone.

The VAP rate improved across the board, with the exception of the rate in the study by Mancebo’s group, which demonstrated no gain. It should be noted that that trial included by far the most severely ill patients, in whom the VAP rate would be expected to be high no matter what was done.

The study conducted by Guerin and coworkers¹¹ is the largest study to date on prone positioning. It attributed a 75% reduction in VAP to prone positioning. That study eclipses the data on head-of-bed elevation when we concider that the control group had key components of the ventilator bundle applied (eg, elevation of head of the bed, postpyloric feeding where feasible)

Patients with blunt chest trauma, who are known to have some of the highest VAP rates, were the focus of another prospective randomized trial. The VAP rate was reduced 68% in trauma patients who were placed in the prone position.¹²

A meta-analysis combining the VAP rates from the 3 aforementioned studies favors proning to reduce VAP rates. Essentially, an odds ratio of 0.79 would indicate that the number of patients needed to be treated is 5. Therefore, for every 5 patients who are flipped over, 1 case of VAP could be avoided. The meta-analysis, further, slightly favored the use of prone when mortality was considered the primary endpoint. When a meta-analysis is performed on other studies that include the most severely ill patients, the benefits of prone ventilation becomes a near slam dunk. Immediately get those patients who have been identified as severely ill in a prone position.

Prone: An Additive to Kinetic Therapy?

If one is not superior, then the question is whether prone ventilation and kinetic together are additive. In a retrospective review that was meant to be a pilot for a prospective study, 61 patients with acute lung injury and/or ARDS were placed on either a supine or prone oscillating bed. Davis and colleagues¹³ found that the kinetic prone group had better P:F ratios of partial pressure of arterial O₂ (PaO₂) to fraction of inspired O₂ (FIO₂) than the supine group by day 5 (243 vs 200). Of course, it is debatable as to whether that should be an endpoint considered the gold standard. There was also a reduction in ventilator days in the prone group, 13 versus 24 days, compared with the supine group. That was a dramatic reduction, although statistically coming in at p=0.12. The prone group also had shorter hospital LOS compared with the supine group (22 vs 40 days). The latter is a trend in LOS reduction, even if it is statistically marginal. Interestingly, the 4 patients whose conditions did not improve with supine kinetic therapy and were subsequently switched to KPT also had significant improvements in PaO₂:FIO₂ ratios.

Probably the most well known of these studies is the one by Gattinoni and coworkers,¹⁴ which had to be shut down early after the interim analysis became available. It was a passive boycott, in that the multi-site investigators were intuitively convinced that their prone patients were doing better and were easier to treat that they refused to enroll patients in the study’s supine control arm.

Kinetic Prone Protocol

On the basis of these data, clinicians at Chandler Regional Hospital (Chandler, AZ) and Banner Desert Medical Center (Mesa, AZ) developed a KPT protocol (Table 2). This evidence-based protocol has been implemented in 3 hospitals, all of which were able to get it up and running within 2 months. Two more hospitals in Denver, Colorado (St. Anthony – Level 1 Trauma & Porter Adventist Hospital) are currently initiating the protocol without difficulty, which speaks to its plug-and-play design.

Table 2. Suggested Protocol for Kinetic Prone Therapy

The KPT protocol recommends putting patients in the prone position for 5 hours at a time with a 62° Kinetic rotation. While the patients are prone, they are put in lateral motion with 10-minute pauses per side. There is no pausing in the middle unless the patient’s hemodynamics must be measured. If the latter is the case, then it is recommended to put a pause in the middle to be consistent in monitoring of hemodynamics. As an alternative, the nursing team is asked to do all of the hemodynamics with the patient in the same position to avoid variations of measurements with regard to patient positioning and the phlebostatic axis point of reference.

Patients are then placed in a supine position for approximately 45 minutes. Although it does not have to be exactly 45 minutes, the goal is to keep the patient prone >20 hours a day. Obviously, patients have to be brought back to supine for standard care, assessments and confirmation of safety measures. Plus, it is important to look at their anterior surface and make sure there are no shear points.

The head of the bed is kept at a maximum elevation of 12° at all times possible. It should be up to maximize the effect of using the abdominal weight to pull down the diaphragm and hold open the lungs, as previously discussed.

Portable chest radiographs are taken daily to ensure stability of lines and tubes. Arterial blood gases are obtained 15 minutes before each flip. The latter is not recommended for all patients if blood conservation in a paramount issue. It has been done at my institutions for data-collection purposes because these patients were the subject of a feasibility trial.

A quick note about informed consent: patients should be informed of all the potential risks. The main one is facial swelling. Other risks include various skin injuries, potential death due to natural disease, and potential blindness. Periorbital edema could cause retinal artery compression, which could result in blindness, although no cases have been reported to date with KPT, case reports exist for prone positioning during prolonged posterior-approach spine surgery.

Finally, hydrocolloid patches are prophylactically placed at high-risk contact points, in areas where patients have the most common shear and pressure injuries, to minimize the risk of developing stage 1 decubitus ulcers. The downside of the patches is that they may attribute to the patients’ average higher temperatures: the Gore-Tex beds, coupled with the patches, act as an insulator.

Briefly, various different methods used to get patients in a prone position are available. The RotoProne (KCI, San Antonio, Tx) is the only automatic bed currently on the market and one employed for the aforementioned protocol. A manual assist devise, the Vollman Proner Positioner (Hill-Rom, Batesville, IN) is available and has varied results. It does preserve neutral alignment, however, so if there is concern about trauma patients who still have not been given C-spine clearance, it is an effective device. Plus, it does make it slightly easier to turn patients even though it still requires 6 or 7 people to do so.

Pilot Study

Thirty-three patients from 2 hospitals have been enrolled in the protocol feasibility pilot study that was mentioned earlier. The study, now in progress, goes one step further than the current evidence-based literature by hypothesizing that an early Kinetic added to Prone therapy protocol will reduce morbidity/mortality and predict the earliest benefit for tracheostomy. This will determine if a tracheostomy should be added to the KPT paradigm.

Thus far, which patients can be weaned from the ventilator at ICU discharge can be predicted by a model at day 3, which is the earliest opportunity to perform a tracheostomy. The patients’ PaO₂:FIO₂ ratios and partial pressure of carbon dioxide (Pco₂) are being charted for the pilot study. The PaO₂:FIO₂ ratios are being collected because the investigators hypothesize that these will be improved, even though the data confirming the importance of PaO₂:FIO₂ ratios has yet to be confirmed. The Pco₂ is being charted because of a retrospective study by Gattinoni that has been completed but not yet reported. In this retrospective study, he returns to the previous one and teases apart the data to see which patients lived and which ones did not. It turns out that patients who are deemed Pco₂ responders had a survival benefit, although the definition of a Pco₂ responder was extremely liberal. That said, if Pco₂ responders are indeed the ones who survived, then it seems as though this is data that should be tracked in subsequent studies.

The bedside application of these preliminary data is as follows: the line defined by the graph in Figure 2 shows a statistically significant breakpoint where patients’ conditions make it apparent whether they will be weaned from the ventilator at ICU discharge. Interestingly, the line did not change over time, specifically 96 hours to ICU discharge. For those patients who could be studied further, the line held, raising a question about the 5-day plateau at which proning is no longer considered beneficial. Unfortunately, patients start to die and the numbers start to drop off further out in this population. Admittedly, this was used as a rescue therapy. Every one of these patients was rescued after traditional ARDS Network ventilation protocols failed.

Although the benefits of KPT appear to plateau in other studies at approximately 5 days, plus or minus 2 days, the pilot study’s numbers started to drop off significantly around day 7. As a routine, however, the patients are returned to a standard ICU bed once their PaO₂:FIO₂ ratio is >250 or >150 if the patient has undergone tracheostomy. Then they are left in the supine position for 4 hours. Every 2 hours, the supine time is lengthened to see if their condition deteriorates.

Figure 2. Margolin pilot study (n = 40).

Frequent arterial blood gas results were plotted with the ratio of partial pressure of arterial oxygen to fraction of inspired oxygen (P/F) over time.

Summary

Patients should be placed prone early to maximize their benefits, especially the most severely ill ones. Keep patients in a prone position for 20 hours a day. As a general rule, automated turning surfacesóas opposed to wedging pillows underneath the patientódecrease secondary injury rates. Most ICUs have some sort of automatic bed, and every major manufacturer makes one that will turn. Use hydrocolloid patches prophylactically to avoid shear/pressure injuries. Finally, prone ventilation should be a part of every institution’s arsenal to reduce VAP in the ICU.

References

Albert RK, Hubmayr RD. The prone position eliminates compression of the lungs by the heart. Am J Respir Crit Care Med. 2000;161:1660–1665.
The SUPPORT principal investigators. A controlled trial to improve care for seriously ill hospitalized patients. The Study to Understand Prognoses and Preferences for Outcomes and Risk of Treatments (SUPPORT). JAMA. 1995;274:1591–1598.
Choi SC, Nelson LD. Kinetic therapy in critically ill patients: Combined results based on meta-analysis. J Crit Care. 1992;7:57–62.
Conrad D, Gardner M. Updated economic implications of the Leapfrog Group Patient Safety Standards: final report to the Leapfrog Group. May 2000. Available at http://www.LeapFrogGroup.org/media/file/Conrad_Updated_Economic_Implications_2_.pdf. Accessed on October 28, 2008.
Ahrens T, Kollef M, Stewart J, et al. Effect of kinetic therapy on pulmonary complications. Am J Crit Care. 2004;13:376–383.
Kirschenbaum L, Azzi E, Sfeir T, et al. Effect of continuous lateral rotational therapy on the prevalence of ventilator-associated pneumonia in patients requiring long-term ventilatory care. Crit Care Med. 2002;30:1983–1986.
Drakulovic MB, Torres A, Bauer TT, et al. Supine body position as a risk factor for nosocomial pneumonia in mechanically ventilated patients: A randomised trial. Lancet. 1999;354: 1851–1858.
Hess DR. Patient positioning and ventilator-associated pneumonia. Respir Care. 2005;50:892–898.
Bryan AC. Pulmonary physiotherapy in the pediatric age group: comment of a devil’s advocate. Am Rev Resp Dis. 1974;110:143–144.
Mancebo J, Fernandez R, Blanch L, et al. A multicenter trial of prolonged prone ventilation in severe acute respiratory distress syndrome. Am J Respir Crit Care Med. 2006;173:1233–1239.
Guerin C, Gaillard S, Lemasson S, et al. Effects of systemic prone positioning in hypoxemic acute respiratory failure. JAMA. 2004;292:2379–2387.
Voggenreiter G, Aufmkolk M, Stiletto RJ, et al. Prone positioning improves oxygenation in post-traumatic lung injuryóa prospective randomized trial. J Trauma. 2005;59:333–341.
Davis JW, Lemaster DM, Moore EC, et al. Prone ventilation in trauma or surgical patients with acute lung injury and adult respiratory distress syndrome: Is it beneficial? J Trauma. 2007;62:1202–1206.
Gattinoni L, Tognoni G, Pesenti A, et al. Effect of prone positioning on the survival of patients with acute respiratory failure. N Engl J Med. 2001;345:568–573.
Papazian L, Gainnier M, Marin V, et al. Comparison of prone positioning and high-frequency oscillatory ventilation in patients with acute respiratory distress syndrome. Crit Care Med. 2005;33:2162–2171.

Self-Assessment

Which best describes kinetic prone therapy?

(a) Putting a patient in a supine position and then rotating the patient >80° each side

(b) Putting a patient in a supine position and then rotating the patient >40° each side

(c) Putting a patient in a prone position and then rotating the patient >80° each side

(d) Putting a patient in a prone position and then rotating the patient >40° each side
Ideally, how long should patients be kept in a prone position?

(a) 5 hours

(b) 11 hours

(c) 12 hours

(d) >20 hours