Show Summary Details
Page of

Prone positioning in the ICU 

Prone positioning in the ICU
Chapter:
Prone positioning in the ICU
Author(s):

Paolo Taccone

and Davide Chiumello

DOI:
10.1093/med/9780199600830.003.0099
Page of

PRINTED FROM OXFORD MEDICINE ONLINE (www.oxfordmedicine.com). © Oxford University Press, 2020. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Medicine Online for personal use (for details see Privacy Policy and Legal Notice).

date: 07 August 2020

Key points

  • Prone positioning optimizes lung recruitment and ventilation–perfusion matching, resulting in an improvement of gas exchange in 70–80% of acute respiratory distress syndrome (ARDS) patients. For this effect, patients with profound life-threatening hypoxaemia may be treated with prone positioning as a rescue manoeuvre.

  • Independently from gas exchange, prone positioning may exert a protective role against ventilator-induced lung injury, reducing the unphysiological stress and strain to which the lung parenchyma is exposed during mechanical ventilation.

  • The current clinical evidences support the use of prone position in the most severe form of ARDS (e.g. PaO2/FiO2 < 150 mmHg, need of high plateau pressure and high positive end-expiratory pressure (PEEP) level, diffuse pulmonary infiltrates), while it should be avoided in less severe patients for its potential adverse effects.

  • Optimal dose of prone positioning is still undetermined. Based on the physiological rationale, we currently suggest to apply prone positioning as long as possible (e.g. >18–20 hours daily), until the resolution of the acute phase of ARDS.

  • While prone positioning does not required any special equipment to be performed, it may significantly increase the risks of potentially life-threatening complications. Therefore, we recommend that prone positioning should be performed with great care by experienced personnel.

Introduction

Prone positioning (also known as ‘proning’, ‘prone manoeuvre’, or ‘prone ventilation’) refers to mechanical ventilation with patients positioned in prone position in contrast of standard supine (flat or semi-recumbent) position. The use of the prone positioning was proposed over 30 years ago as a means to improve arterial oxygenation in patients with acute respiratory distress syndrome (ARDS) [1]‌. Since then, extensive physiological research has been conducted to explore the possible mechanisms underlying the observed improvement in gas exchange, which involve changes in the distribution of both ventilation and pulmonary blood flow. Furthermore, it has been shown that, independently of gas exchange, prone positioning may reduce the harm of mechanical ventilation [2,3], which is known to adversely impact patient survival. In this chapter, we will summarize the physiological effect of prone positioning, as well as the clinical evidences supporting its use to reduce mortality in patients with ARDS [4,5].

Physiological effects of prone positioning

The effects of prone positioning on gas exchange results from a redistribution of lung ventilation and pulmonary perfusion. Indeed, in any body position, regional lung ventilation and pulmonary blood flow are influenced by the gravitational field of the earth. When we apply prone positioning to a patients, we reverse the vector of this gravitational force, with major consequences on inflation and perfusion distribution. As a secondary and less relevant effect, prone positioning also promotes postural drainage of secretions from the tracheobronchial tree, possibly ameliorating regional lung ventilation.

Distribution of ventilation

The distribution of alveolar lung inflation is different in prone and supine position [6]‌. In a normal subject lying flat in supine position, the pleural pressure, that is the determinant of alveolar dimension, results greater in the dorsal posterior region that in the non-dependent sternal region, mainly because of the hydrostatic pressure imposed by the weight of the lung tissue. Therefore, the non-dependent alveoli, located near the sternum, are more inflated than the dependent one in the dorsal region (see Fig. 99.1). In patients with ARDS, this inflation heterogeneity is exaggerated by the dramatic increase in lung weight due to the widespread inflammatory lung oedema. In this condition, the increase in pleural pressure gradient leads to an over inflation of the non-dependent region and a compression atelectasis of the dependent region. When the patient is turned in prone position, the inflation gradient is reversed, and the regional alveolar size is greater in dorsal regions and lower in ventral regions. Once again, the dependent lung regions (ventral) are collapsed [7]. Of note, in prone position the pleural pressure gradient is significantly decreased compared with supine position, resulting in a more homogeneous distribution of alveolar inflation (see Fig. 99.1). Accordingly, a reduced amount of collapsed lung parenchyma (i.e. more lung recruitment) and decreased alveolar over inflation is observed with prone positioning.

Fig. 99.1 Lung inflation (expressed as gas/tissue ratio measured by CT scan) for each lung section along the ventrodorsal axis (on the x-axis, level 0 represents the most non-dependent lung section, while level 10 represents the most dependent one). Data have been collected on healthy subjects (open symbols) and ARDS patients (closed symbols). The degree of alveolar inflation reflected by the gas/tissue ratio is that observed in the supine (square symbols) and prone position (dot symbols), respectively.

Fig. 99.1 Lung inflation (expressed as gas/tissue ratio measured by CT scan) for each lung section along the ventrodorsal axis (on the x-axis, level 0 represents the most non-dependent lung section, while level 10 represents the most dependent one). Data have been collected on healthy subjects (open symbols) and ARDS patients (closed symbols). The degree of alveolar inflation reflected by the gas/tissue ratio is that observed in the supine (square symbols) and prone position (dot symbols), respectively.

Reproduced from Gattinoni L et al., ‘Prone positioning in acute respiratory distress syndrome’. In: Tobin MJ, Principles and Practice of Mechanical Ventilation, © 2012 McGraw-Hill Education.

Another phenomenon has been recently emphasized to contribute to the redistribution of lung inflation during prone positioning, which is related to the position of the heart in the thorax. Indeed, while in supine position the cardiac mass has a direct compressive effect on the posterior regions of the lung (particularly the left lower lobe), in prone position the lungs are relieved from the heart weight [87], as this lies directly on the sternum. This effect may be particularly pronounced in patient with cardiomegaly.

Distribution of perfusion

In normal subject, pulmonary blood flow progressively increases from the non-dependent to the dependent region of the lung, following gravitational distribution. In ARDS patients, factors other than gravity influence regional distribution of lung perfusion force, as hypoxic pulmonary vasoconstriction and extrinsic compression of the vessel by mechanical ventilation. Experimental studies have confirmed that in supine position ARDS leads to a high shunt fraction in the dorsal region (i.e. high perfusion and poor aeration). Interestingly, when the prone position is applied, perfusion distribution is less gravity-dependent than in the supine position, and a substantial fraction of pulmonary perfusion is maintained in the dorsal region. As this phenomenon is coupled to a recruitment of the dorsal region with prone position, the reopened alveoli continues to receive the majority of blood flow, resulting in a shunt reduction and a better matched ventilation/perfusion ratio [9]‌.

Prone positioning and clinical outcome

Effect on oxygenation

Based on the results of many observational studies, as well as randomized controlled trials (RCTs), there is wide agreement that prone positioning increases arterial oxygen tension in most of the patients with ARDS. However, the degree of the response was variable, ranging from great improvement, to no change, and even a significant deterioration in a small fraction of patients. Among patients whose oxygenation improves, this improvement is usually progressive while in the prone position, showing a time-dependent effect. Furthermore, when returning to a supine position after a period in a prone position, some patients maintain an oxygenation benefit for hours, while other rapidly return to their basal supine oxygenation. Finally, the effect of prone positioning may change over the course of ARDS, usually decreasing in its benefit when lung pathology progresses from the oedematous phase to the fibrotic phase (i.e. after 7–10 days).

In conclusion, the short- and long-term oxygenation response while in the prone position are highly variable, probably because the individual response is strictly dependent on patients’ underlying pathophysiological status. Notably, while several clinical variables have been investigated as possible predictors of this response, none of them have shown sufficient accuracy to be considered reliable at the bedside.

Effect on mortality

In the past, it has been suggested that prone positioning improves oxygenation, but in recent years there has been a progressive recognition that the prone position may decrease the stress and strain to which the lung parenchyma is exposed during mechanical ventilation [2,3]. Therefore, the possible survival advantage of prone positioning should be independent of oxygenation changes, which were constantly demonstrated to be uncorrelated with outcome, but it may be related to a decrease in the danger associated with mechanical ventilation (ventilator induced lung injury).

To date, five high-quality RCTs have addressed whether prone positioning might decrease the mortality of adult patients with ARDS:

  • The Prone-Supine study group [10], enrolled 304 patients presenting with acute lung injury (ALI) or ARDS. This study investigates prone positioning applied for a mean of 7 hours daily, for a maximal period of 10 days (mean 4.7 days of treatment). Other relevant co-treatment, as mechanical ventilation settings were not protocolized. Although no effect on 6-month mortality was found in the overall study population, a post hoc analysis showed a trend towards reduced short-term mortality in the subgroup of patients which presented the most severe form of ARDS.

  • The Guerin et al. study [11]: consisted of a larger population of 802 patients with hypoxaemic acute respiratory failure from various aetiologies (only about half of them fulfilled the ALI/ARDS criteria). The treatment protocol was 9 hours of prone positioning per day, applied until clinical criteria of improvement were matched (mean 4.1 days of treatment). No effect on primary outcome was found, and prone positioning was associated with a significant increase in adverse events.

  • The Mancebo et al. study [12]: randomized 136 ARDS with diffuse radiologic pulmonary infiltrates. This study investigates a prolonged prone positioning strategy (mean 17 hours daily for 10.1 days of treatment), and included a lung protective mechanical ventilation protocol in both study arms. The findings of the study were negative, but there was a not statistically significant trend in intensive care unit (ICU) mortality reduction of about 15% in the treatment arm.

  • The Prone-Supine study II [13]: enrolled 342 ARDS patients, using a prolonged prone positioning protocol (mean 18 hours daily for 8.3 days of treatment) and a protocolized protective mechanical ventilation strategy. Based on the result of the previous studies, patients were stratified at enrolment according to the severity of their hypoxaemia (severe hypoxaemia defined as PaO2/FiO2 ratio below 100). No significant effect on mortality was found in the overall population, while a trend in 6-month mortality reduction of about 10% was demonstrated in the most severely hypoxaemic patients.

  • The PROSEVA study [14]: randomized 466 severe ARDS patients with a PaO2/FiO2 ratio lower than 150 with a positive end expiratory pressure (PEEP) higher than 5 cmH2O to prone positioning for at least 16 hours or to supine position. The 28-day mortality rate was 16% in the Prone group compared with 32.8% in the supine group (P < 0.001). Patients were ventilated in the prone position for 73% of the 22.334 patient hours from the enrolment to the last session.

Notably, all these trials have different potential methodological bias, such as inadequate patient selection, underpowered population, possible suboptimal treatment strategy, and lack of standardization for relevant co-treatment (e.g. protective low tidal volume ventilation). However, the last trial published finally showed a definitive evidence supporting that prone positioning has a beneficial effect on survival in the most severely ill patients. Accordingly, two patients-level meta-analyses with different inclusion criteria were conducted in collaboration with trialists of the first four trials [4,5], aiming to study the interaction between the severity of hypoxaemia and the response to prone positioning. Both studies confirmed a survival advantage of prone positioning of about 10% in the most severely hypoxemic patients, with no significant heterogeneity among trials [5]‌. On the contrary, no effect was observed in patients with less severe hypoxaemia, discouraging the routine use of prone positioning in this subgroup of patients (see Fig. 99.2).

Fig. 99.2 Kaplan–Meier estimates of survival rates of the prone (solid line) and supine (dashed line) patients from a patient-level meta-analysis of the four largest RCTs investigating the effects of prone positioning on mortality [10,11,12,13]: (a) entire ARF population, (b) moderately hypoxaemic patients (PaO2/FiO2 100–200 mmHg at baseline, and (c) severely hypoxaemic patients (i.e. PaO2/FiO2 < 100 mmHg.
Fig. 99.2 Kaplan–Meier estimates of survival rates of the prone (solid line) and supine (dashed line) patients from a patient-level meta-analysis of the four largest RCTs investigating the effects of prone positioning on mortality [10,11,12,13]: (a) entire ARF population, (b) moderately hypoxaemic patients (PaO2/FiO2 100–200 mmHg at baseline, and (c) severely hypoxaemic patients (i.e. PaO2/FiO2 < 100 mmHg.
Fig. 99.2 Kaplan–Meier estimates of survival rates of the prone (solid line) and supine (dashed line) patients from a patient-level meta-analysis of the four largest RCTs investigating the effects of prone positioning on mortality [10,11,12,13]: (a) entire ARF population, (b) moderately hypoxaemic patients (PaO2/FiO2 100–200 mmHg at baseline, and (c) severely hypoxaemic patients (i.e. PaO2/FiO2 < 100 mmHg.

Fig. 99.2 Kaplan–Meier estimates of survival rates of the prone (solid line) and supine (dashed line) patients from a patient-level meta-analysis of the four largest RCTs investigating the effects of prone positioning on mortality [10,11,12,13]: (a) entire ARF population, (b) moderately hypoxaemic patients (PaO2/FiO2 100–200 mmHg at baseline, and (c) severely hypoxaemic patients (i.e. PaO2/FiO2 < 100 mmHg.

Reprinted by permission of Edizioni Minerva Medica from Minerva Anestesiol 2010; 76, 448–54.

The hypothesis that prone positioning may be more effective in the most severe patients has a strong pathophysiological rationale. Indeed, these patients have been extensively demonstrated to present greater amount of pulmonary oedema, more widespread alveolar collapse and greater lung recruitability [18], and prone positioning exert its lung-protective effect mainly recruiting the collapsed regions of the lung.

Other outcomes

Prone positioning has been hypothesized to reduce the incidence of ventilator-associated pneumonia through improved reduction of secretion. Although some encouraging positive results, most trials evaluating VAP as an outcome were flawed by major limitations, and this finding should be taken cautiously [5]‌.

Application of prone positioning

Patients selection

We believe that available clinical and preclinical data support the use of prone positioning in the management of patients with the most severe form of ARDS. In contrast, given the potentially harmful effects, prone positioning should not be routinely used in patients with less severe ARDS [4,5].

A practical question facing the physician at the bedside is which level of hypoxaemia should be used to identify ARDS patients that may possible benefit from prone positioning. Some RCT investigators proposed a threshold of PaO2/FiO2 ratio below 100 mmHg (with a PEEP higher that 5 cmH2O) [10,13], others 150 mmHg [14]. However, although the PaO2/FiO2 ratio is correlated to the severity of the disease process, it is also highly dependent to other confounding factors (e.g. ventilatory strategy, PEEP response, haemodynamic status, fluid balance, etc.), and its reliability and reproducibility as a single indication criterion for prone treatment is questionable. Therefore, we currently suggest that clinicians should evaluate ARDS severity using a PaO2/FiO2 threshold (e.g. below 150) preferentially in conjunction with other markers of severity, as the direct/indirect evidence of high recruitability of the lungs (e.g. diffuse infiltrates at CXR or CT imaging), a severe impairment of respiratory mechanics (e.g. high plateau pressure), and a rapidly progressive deterioration of gas exchange unresponsive to conventional ventilation. Furthermore, a lack of oxygenation improvement should not be used as an absolute criterion to discontinue prone positioning. Indeed, as oxygenation response may depends to phenomenon unrelated to lung recruitment (as pulmonary blood flow diversion) [19], it is not a reliable marker of lung protection during prone positioning.

There are very few absolute contraindications to prone positioning, as spinal instability and unmonitored increased intracranial pressure. Other conditions should be identify as relative contraindication, as open abdominal wounds, multiple trauma with unstabilized fracture, pregnancy, severe haemodynamic instability, and high dependency on airway and vascular access (e.g. extracorporeal membrane oxygenation support).

Positioning and timing of treatment

Prone positioning does not required any special equipment, and it can be safely performed manually by 3–5 specifically-trained health care personnel. Recently, some commercially-available beds have been specifically developed to perform the turning manoeuvre and/or maintain the positioning, but the reduction of workload and the possible advantage of procedure standardization must be weighed against the costs of these rather pricy devices.

In our clinical practice we start prone positioning as soon as the patients is diagnosed as having a severe form of ARDS. It is usual to wait a few hours to allow initial stabilization and perform any diagnostic procedures needed, but if the patient presents with an urgent life-threatening hypoxaemia, which is unresponsive to a highly-supporting mechanical ventilation (e.g. PaO2 < 55 mmHg with PEEP > 20 cmH2O and FiO2 100%), prone positioning should be immediately assumed as a rescue manoeuvre.

The optimal daily duration of prone positioning is still unknown [5]‌. The final RCTs [12,13] applied a longer time of prone positioning compared with early trials [10,11] (i.e. 17–18 versus 7–9 hours/day), and the PROSEVA study, which used a prolonged prone position clearly showed a significant reduction in mortality [14].

Furthermore, the optimal timing and weaning criteria from prone positioning remain undetermined. Some trials suggested a shorter ‘acute phase’ protocol [10,11,14], while others prolonged the application of the treatment until the final phases of weaning from mechanical ventilation [12,13]. In the absence of certain information, based on the physiological rationale and clinical data available, currently suggested that a daily prone position be applied for as long as possible (e.g. >18–20 hours daily), until the resolution of the acute phase of ARDS (i.e. <7 days).

Finally, ventilation of patients in the prone positioning according to a lung-protective ventilation strategy is strongly recommended. Moreover, the use of prone positioning has been proposed in association to other non-conventional treatments [14,20] (e.g. inhaled nitric oxide, high-frequency oscillatory ventilation, extracorporeal membrane oxygenation), but the benefit of these combinations remained unpredictable and their use needs to be limited to selected patients in highly specialized centres.

Adverse events

Several side effects have been associated to the use of prone positioning, with some between-report differences in actual occurrence and incidence [10,11,12,13,14]. Some complications of prone positioning are reported as uncommon, but they may be dramatic and potentially life-threatening, especially in severely unstable patients (e.g. inadvertent extubation, displacement of thoracotomy tubes or extracorporeal membrane oxygenation access, major arrhythmias). Other less severe and more common adverse events include the displacement of vascular accesses, need for increased sedation or muscle relaxants, airway obstruction, transient desaturation, hypotension or increased use of vasopressor, and vomiting. Furthermore, dependent facial oedema, pressure ulcers, and some rare cases of nerve compression or retinal damage may occur, and they must be prevented or minimized by careful positioning and the use of adequate soft-padding.

Of note, one of the RCT published a demonstrated a significant increase in the rate of adverse events associated to a prolonged strategy of prone positioning [13]. However, we believe these increased risks may be partially explained by the increased frequency of turning manoeuvres required by the study protocol (i.e., every 4 hours). Therefore, we currently suggest reducing the number of turning as low as possible.

Conclusion

In conclusion, in order to minimize the potential risks, we suggest that prone positioning should be applied only by specifically-trained personnel with adequate experience in its use. A high level of attention of ICU staff is mandatory, especially during the turning manoeuvre, with maximal effort to prevent, or promptly recognize and correct, any possible major complication.

References

1. Piehl MA and Brown RS. (1976). Use of extreme position changes in acute respiratory failure. Critical Care Medicine, 4(1), 13–14.Find this resource:

2. Valenza F, Guglielmi M, Maffioletti M, et al. (2005). Prone position delays the progression of ventilator-induced lung injury in rats: does lung strain distribution play a role? Critical Care Medicine, 33(2), 361–7.Find this resource:

3. Broccard A, Shapiro RS, Schmitz LL, Adams AB, Nahum A, and Marini JJ. (2000). Prone positioning attenuates and redistributes ventilator-induced lung injury in dogs. Critical Care Medicine, 28(2), 295–303.Find this resource:

4. Gattinoni L, Carlesso E, Taccone P, Polli F, Guerin C, and Mancebo J. (2010). Prone positioning improves survival in severe ARDS: a pathophysiologic review and individual patient meta-analysis. Minerva Anestesiologica, 76(6), 448–54.Find this resource:

5. Sud S, Friedrich JO, Taccone P, et al. (2010). Prone ventilation reduces mortality in patients with acute respiratory failure and severe hypoxemia: systematic review and meta-analysis. Intensive Care Medicine, 36(4), 585–99.Find this resource:

6. Gattinoni L, Taccone P, Valenza F, Pelosi P, and Mascheroni D. (2012). Prone positioning in acute respiratory distress syndrome. In: Tobin MJ (ed.) Principles and Practice of Mechanical Ventilation , 699–726. New York, NY: McGraw-Hill.Find this resource:

7. Gattinoni L, Pelosi P, Vitale G, Pesenti A, D’Andrea L, and Mascheroni D. (1991). Body position changes redistribute lung computed-tomographic density in patients with acute respiratory failure. Anesthesiology, 74(1), 15–23.Find this resource:

8. Albert RK and Hubmayr RD. (2000). The prone position eliminates compression of the lungs by the heart. American Journal of Respiratory and Critical Care Medicine, 161(5), 1660–5.Find this resource:

9. Richter T, Bellani G, Scott Harris R, et al. (2005). Effect of prone position on regional shunt, aeration, and perfusion in experimental acute lung injury. American Journal of Respiratory and Critical Care Medicine, 172(4), 480–7.Find this resource:

10. Gattinoni L, Tognoni G, Pesenti A, et al. (2001). Effect of prone positioning on the survival of patients with acute respiratory failure. New England Journal of Medicine, 345(8), 568–73.Find this resource:

11. Guerin C, Gaillard S, Lemasson S, et al. (2004). Effects of systematic prone positioning in hypoxemic acute respiratory failure: a randomized controlled trial. Journal of the American Medical Association, 292(19), 2379–87.Find this resource:

12. Mancebo J, Fernandez R, Blanch L, et al. (2006). A multicenter trial of prolonged prone ventilation in severe acute respiratory distress syndrome. American Journal of Respiratory and Critical Care Medicine, 173(11), 1233–9.Find this resource:

13. Taccone P, Pesenti A, Latini R, et al. (2009). Prone positioning in patients with moderate and severe acute respiratory distress syndrome: a randomized controlled trial. Journal of the American Medical Association, 302(18), 1977–84.Find this resource:

14. Guerin C, Reignier J, Richard JC et al. (2013). Prone positioning in severe acute respiratory distress syndrome. New England Journal of Medicine, 368, 2159–68.Find this resource:

15. Sud S, Sud M, Friedrich JO, and Adhikari NKJ. (2008). Effect of mechanical ventilation in the prone positioning on clinical outcomes in patients with acute hypoxemic respiratory failure: a systematic review and meta-analysis. Canadian Medical Association Journal, 178(9), 1153–61.Find this resource:

16. Abroug F, Ouanes-Besbes L, Elatrous S, and Brochard L. (2008). The effect of prone positioning in acute respiratory distress syndrome or acute lung injury: a meta-analysis. Areas of uncertainty and recommendations for research. Intensive Care Medicine, 34(6), 1002–11.Find this resource:

17. Alsaghir AH and Martin CM. (2008). Effect of prone positioning in patients with acute respiratory distress syndrome: a meta-analysis. Critical Care Medicine, 36(2), 603–9.Find this resource:

18. Gattinoni L, Caironi P, Cressoni M, et al. (2006). Lung recruitment in patients with the acute respiratory distress syndrome. New England Journal of Medicine, 354(17), 1775–86.Find this resource:

19. Gattinoni L, Vagginelli F, Carlesso E, et al. (2003). Decrease in PaCO2 with prone position is predictive of improved outcome in acute respiratory distress syndrome. Critical Care Medicine, 31(12), 2727–33.Find this resource:

20. Pipeling MR and Fan E. (2010). Therapies for refractory hypoxemia in acute respiratory distress syndrome. Journal of the American Medical Association, 304(22), 2521–7.Find this resource: