Prone Position in Acute Respiratory Failure
By Francisco Baigorri, MD, and Lluis Blanch, MD
Introduction: Why consider the prone position in respiratory failure?
Bipedalism seems to have been one of the earliest of the major hominine characteristics to evolve, several millions years ago. Yet it seems that human evolution has not yet reached the supine positionat least as far as our patients with acute respiratory failure are concerned. Indeed, in comparison with the prone position, the supine position seems to be physiologically disadvantageous for the lung.
Gravitational distribution of pleural pressure is much more uniform in the prone than in the supine position.1,2 Hence, regional inflation is more homogeneously distributed in the prone position, and less alveolar collapse occurs at end-expiration. Caudal displacement of the dorsal diaphragm is a major factor involved in the change of ventilation in the prone position.3 On the other hand, regional perfusion is relatively unaffected by a change from supine to prone position.4 Therefore, the prone position should be associated with an improvement of ventilation-perfusion heterogeneity. This improvement has been shown in animals with acute lung injury induced by oleic acid5 and in humans with the acute respiratory distress syndromt (ARDS).6 Indeed, one recent study in animals with oleic acid-induced lung injury and ventilated with high tidal volumes and positive end-expiratory pressure (PEEP) showed that animals in the prone position undergo less extensive lung histologic change than those in the supine position.7
Consequently, turning patients from the supine to the prone position has been proposed as a useful supportive therapy that can improve oxygenation in many patients with ARDS. Clinical studies in ARDS patients have reported that changing to the prone position improves oxygenation in 60-70% of patients8-10 and has no deleterious effect on hemodynamics. Almost 50% of these patients maintained their improvement when returned supine. Moreover, the prone position was well tolerated and was rarely associated with significant impairments in PaO2 in non-responders. In addition, pulmonary secretions were more easily aspirated after performing the turn. In one study, the use of prone positioning, combined with pressure-limited ventilation in severe ARDS, has been associated with a mortality rate lower than that predicted by the APACHE II score.10
However, although the earliest clinical studies showing an improvement of PaO2 in patients with ARDS were reported more than 20 years ago,11,12 there remains considerable controversy about the appropriate use of this supportive therapy, and also in regard to outcomes. Further randomized trials are needed to elucidate these aspects. Nonetheless, it is useful to have a simple, efficient method to improve gas exchange in patients presenting with acute respiratory failure that is unresponsive to conventional therapy.9 Here, we will discuss practical aspects regarding the use of prone positioning in the light of our current knowledge.
Which patients are best suited to trials with prone positioning?
In our institution, the prone position is considered as supportive treatment when inspired oxygen requirements are higher than 60% in spite of PEEP therapy instituted in an attempt to reduce the inspired oxygen fraction (FiO2) and/or the level of PEEP. Prone positioning is now being used sooner than in the past, as we have gradually become more familiar with the technique and have verified the absence of complications.13
Seventy percent of patients showed an improvement in oxygenation, as measured by an increase in the PaO2/FiO2 ratio of more than 15%.14 Responding patients had significantly lower PaO2/FiO2 ratios and higher PaCO2 values prior to proning than those who did not respond to the position change. However, other authors have associated the lack of response with a number of more severe clinical variables, such as higher values for pulmonary artery pressures8 or lower PaO2/FiO2 ratio and higher PEEP.9 We have reported elsewhere14 that the time elapsed after onset of ARDS before prone positioning was longer in nonresponders (33 ± 48 vs 12 ± 16 days).
In contrast, Chatte and colleagues9 found that the time spent using ventilation before the prone trial was significantly shorter in nonresponders (6 ± 8 vs 8 ± 10 days). Vollman and Bander even suggested that nonresponders have additional lung pathologies or more segmental lung pathologies. Thus, the mechanisms underlying the lack of response to the position change remain unclarified, and we do not have data that allow us to anticipate which patients will benefit from prone positioning.
In any case, only a few nonresponders fail to tolerate prone positioning and need to be rapidly returned to supine;9so testing the effect of prone position in patients with ARDS requiring high FiO2 or PEEP levels is not formally contraindicated, except in the presence of unstable fractures or uncontrolled hemodynamics.
How long should patients be left prone?
No clear strategy has been established for alternating prone and supine periods. Most of the improvement in PaO2 occurs rapidlywithin a few minutes after turning6,9,15and it has been shown that more than 50% of responding patients maintain their improvement when returned supine after the first four hours of the prone trial.9 Thus, in the case of a positive response, a persistent or nonpersistent effect when returning to supine can be evaluated after a few hours. Prone positioning can be repeated, depending on the evolution of the PaO2/FiO2 ratio. Moreover, prone-dependent patients can spend long periods in this position; some of our patients have remained in the prone position for more than 24 hours without major side effects. Only minor complications such as periorbital and conjunctival edema were detected.13
Tips that may help clinicians carrying out prone positioning for the first time
Some argue that turning patients prone is difficult and risky, but this concern is not supported either by data in literature,9,10 nor by our experience.13,14 Probably, as recently noted by Albert,16 those who argue most vociferously about the difficulties of turning patients prone seem to be the ones with the least experience in doing so. However, turning does require forethought.9 In our unit, the turn is carried out by a team of four or five people. Each member of the team has a definite role, and the team is led by a physician who is in charge of the patient’s airway during the procedure. Under these premises, we have successfully turned patients prone even in life-threatening conditions with severe hypoxemia, despite an FiO2 of 1.0, high PEEP, and administration of inhaled nitric oxide, who also required invasive hemodynamic monitoring and vasoactive drugs. The dramatic improvement observed in most of these extreme cases has been encouraging for everyone involved in their care.
Figure
Intubated Patient in Prone Position
Patient lying prone on the bed with pillows supporting the head, upper chest, and pelvic region, leaving the abdomen as free of restriction as possible.
Sedation is usual in these most severe cases, and it is commonly continued in the prone position. Caution should be taken to avoid stiff support. One special point of discussion is whether the abdomen can rest directly on the bed or should be maintained free of restriction. Better results were obtained when the abdominal mass could more freely expand downward,9 as diagrammed in the Figure above. Vollman and Bander proposed a prone positioning device, which consists of a portable frame comprised of four plastic/foam pieces supporting the forehead, chin, upper chest, and pelvic region. The chest piece is aligned to rest between the clavicle and sixth rib. The pelvic piece is adjusted to rest one-half inch above the iliac crest. Pictures of this device can be helpful to those unfamiliar with prone positioning.
Although no special bed is required, the availability of devices such as the one proposed by Vollman and Bander, or beds specifically designed for prone positioning, may reduce the number of people required for the position change.
Conclusion
Turning critically ill patients from supine to prone position is an efficient and simple way of improving gas exchange in patients presenting with acute respiratory failure. Since prone positioning seems physiologically advantageous for the lung if compared with the supine position, early, widespread use of this method can be expected in the near future. However, several questions regarding this postural change remain unanswered, including its influence on outcomes, the mechanisms that determine a positive or negative response to the prone trial, and the most appropriate time frequency of the turns. Until we obtain answers to these questions, the prone position should probably be considered as an adjunctive therapy to be started when conventional treatments have failed. (Dr. Blanch is Staff Intensivist, Hospital Parc Tauli, Univ Autonoma de Barcelona, Spain.)
References
1. Yang QH, Lai-Fook SJ. J Appl Physiol 1991;71:76-82.
2. Mutoh T, et al. Am Rev Respir Dis 1992;300-306.
3. Krayer S, et al. Anesthesiology 1989;70:891-898.
4. Wiener CM, et al. J Appl Physiol 1990;68:1386-1392.
5. Lamm WJE, et al. Am J Respir Crit Care Med 1994; 150:184-193.
6. Pappert MD, et al. Chest 1994;106:1511-1516.
7. Broccard AF, et al. Crit Care Med 1997;25:16-27.
8. Vollman KM, Bander JJ. Intensive Care Med 1996;22: 1105-1111.
9. Chatte G, et al. Am J Respir Crit Care Med 1997;155: 473-478.
10. Stocker R, et al. Chest 1997;111:1008-1017.
11. Piehl MA, Brown RS. Crit Care Med 1976;4:13-14.
12. Douglas WW, et al. Am Rev Respir Dis 1977;115: 559-566.
13. Hernandez E, et al. Intensive Care Med 1996;22:420S.
14. Blanch L, et al. Intensive Care Med 1996;22:419S.
15. Langer M, et al. Chest 1988;94:103-107.
16. Albert RK. Am J Respir Crit Care Med 1997;155: 393-394.