MATHEMATICS OF PRONE POSITION AND IMPLICATION IN ARDS

"It appears that pulmonary mechanics is evolutionary designed and optimised for prone position, which is the normal posture in quadrupeds"

 

Prone position is used in operation rooms for a long time for surgical procedures on dorsal aspect of body. It has been acknowledged that respiratory and hemodynamic instability may result because of compression of chest and abdomen in prone position, which may be relieved by supporting upper chest and pelvis with padding, to facilitate free movement of anterior abdominal wall.

 

Beneficial effect of prone position in respiratory mechanics was first proposed by Bryan in 1974. Bryan studied the effect of sedation and paralysis on diaphragm, and suggested that prone position improve dorsal lung expansion and oxygenation¹. Few years later  Piehl and Douglas separately reported that prone position dramatically improved oxygenation in ARDS patients^2,3.

 

But, it had to take one more decade, to understand the mechanism of prone position in ARDS. 1986 was the turning point in the history of ARDS, when Maunder and colleague challenged the long held belief that ARDS is a homogenous process, as seen in X ray images. CT scan images of ARDS lung revealed that densities are heterogeneously distributed. Gattinoni et al showed that alveolar densities are mainly located in dependent regions of lung, with normal compliant baby lung in non-dependent regions.

To understand the prone position physiology, we have to first delve into the anatomy and physiology of respiratory system as well as pulmonary mechanics of ventilation.

 

ANATOMY AND PHYSIOLOGY OF RESPIRATORY SYSTEM- Chest and abdominal compartments are separated by a thin and compliant diaphragm. Chest wall is more compliant ventrally than dorsally, because of greater freedom of movement of ribcage anteriorly. Abdomen has a compliant ventral wall. As hydrostatic pressure in the abdominal compartment is higher than in chest cavity by a factor of five (1.0 vs 0.2 cmH₂O/cm height, respectively)¹, a passive diaphragm is more compliant ventrally than dorsally. Effect of intraabdominal pressure on diaphragmatic compliance becomes exaggerated in acute illness like sepsis, pancreatitis, trauma with fluid resuscitation.

Anatomically conical shaped (in both longitudinal and transverse plane) elastic lung is lodged within cylindrical shaped rigid chest wall. As a consequence, dorsal and caudal regions of chest cavity have higher lung tissue mass compared to ventral and apical part. In supine position, nearly 20% of lung tissue is oriented in ventral plane, compared to 50% in dorsal plane (below the level of heart)⁴.

Alveolar Ventilation is heterogeneously distributed in different lung regions along both gravitational and iso-gravitational plane. Ventilation of regional alveolar units is determined by pulmonary mechanics and fractal geometry of bronchial airways. 

Regional ventilation of lung is governed by three factor, alveolar volume at end expiration (FRC), alveolar compliance and distending pressure at end inspiration (transpulmonary pressure at end inspiration). These lung mechanics are the factor that cause unequal distribution of ventilation in vertical plane.

Regional alveolar volume at end expiration, in normal lung, is heterogeneously distributed along gravitational plane, with larger alveoli at the top compressing the smaller alveoli towards bottom. Regional alveolar compliance also follow the same trend as alveolar volume. Transpulmonary pressure at end inspiration, is determined by pleural pressure in spontaneous breathing, while positive airway pressure in mechanical ventilation.

Fractal geometry of airways influences regional distending pressure at end inspiration by manipulating gas flow along subsequent generations of airways. It is mainly responsible for ventilation heterogeneity in iso-gravitational plane.

Pulmonary perfusion is also determined by gravity and fractal geometry of pulmonary vessels. Like ventilation, perfusion is heterogeneously distributed both in vertical and horizontal plane. But, unlike ventilation it is highest in dorsal and caudal region, regardless of the posture. In fact, only about 1-25% of pulmonary perfusion is influenced by gravitational forces⁵.

 

 

 

POLMONARY MECHANICS IN SUPINE POSITION

As conical and elastic lung is lodged within rigid and cylindrical chest wall, larger lung tissue is present in dorsal and caudal regions than ventral and apical. Because of the effect of shape matching, pleural pressure tends to become less negative in dorsal regions of lung than ventral. The resulting vertical gradient in pleural pressure produces higher transpulmonary pressure and alveolar volume towards ventral regions and lesser transpulmonary pressure and alveolar volume towards ventral regions.

Simultaneously, gravity forces the conical lung to assume its original shape, resulting in ventral alveolar units to sag and compress dorsal alveolar units, further reducing the size of dependent alveolar units in supine position. Therefore, the combined effect of shape matching and gravity produces decremental vertical gradient in regional alveolar volume. In pressure volume relationship, the compressed and smaller alveoli at the dependent regions are relatively more compliant, than larger alveoli at the non-dependent regions. In terms of pulmonary mechanics, alveolar stress and strain is heterogeneously distributed in normal lung, with higher stress-strain in ventral region than in dorsal region.

Transpulmonary pressure at end inspiratory is determined by end inspiratory alveolar pressure and pleural pressure. Transpulmonary pressure at end inspiration is regulated differently in spontaneous ventilation and positive pressure ventilation. Primary determinant of transpulmonary pressure in spontaneous breathing is pleural pressure, which is manipulated by inspiratory muscles of respiration, while alveolar pressure remains constant (atmospheric). In positive pressure ventilation, alveolar pressure (airway plateau pressure) is the manipulated variable, while pleural pressure remains constant.

In spontaneous breathing, diaphragm is the primary muscle of inspiration, which contributes to 72% of the tidal inspiration. Because of it curvature, diaphragm expands more in dorso-caudal region, generating relatively more negative pleural pressure in these lung regions⁶. Consequently, transpulmonary pressure at end inspiration is higher in dependent dorsal and caudal region of the lung, than non-dependent regions of lung.

Therefore, smaller alveolar volume at end expiration, better alveolar compliance and higher transpulmonary pressure at end inspiration toward dorsal and caudal regions of lung, produces preferential distribution of ventilation towards dorsal and caudal regions of lung in spontaneous breathing.

In terms of pulmonary mechanics, spontaneous breathing attenuates the heterogeneity in regional alveolar stress-strain, making it more uniform throughout the lung.

In mechanical ventilation, positive pressure displaces the passive diaphragm more in the more compliant non-dependent ventral region. Also, greater freedom of movement of ribcage in ventral than dorsal chest wall, allows preferential distribution of ventilation in non-dependent ventral regions of lung in supine position. In a sedated, paralyzed patient, rib cage expansion during mechanical ventilation, increases from 40% to 72% of total chest expansion, in comparison to spontaneous breathing⁷.

Thus, in positive pressure ventilation, larger and less compliant non-dependent alveoli are over inflated, while smaller and more compliant dependent alveoli remain underinflated or collapsed. in other words, alveoli in non-dependent regions experience greater stress-strain than those at dependent regions. Increased heterogeneity in alveolar size in ARDS further exaggerates this effect, and consequent risk of VILI.

Thus, in spontaneous breathing, ventilation is preferentially distributed to dorsal regions, while in mechanical ventilation, ventilation is preferentially distributed to ventral regions of lung, in supine position.

In supine position, heart compresses the part of lung posterior to it. Though this has little effect in a healthy lung during spontaneous breathing, it becomes significant in pathological conditions like ARDS and intra-abdominal hypertension.

Therefore, in supine position during spontaneous breathing, both ventilation and perfusion are preferentially distributed towards dorsal dependent region of lung, with consequent lesser V/Q mismatching and stress-strain. But, during mechanical ventilation, ventilation is preferentially distributed towards non-dependent ventral regions, while perfusion is distributed towards dorsal regions, with resulting increased V/Q mismatch and exaggerated stress-strain heterogeneity.

 

PULMONARY MECHANICS IN PRONE POSITION

In prone position, effect of shape matching and gravity on pleural pressure, try to oppose each other. Consequently, pleural pressure, transpulmonary pressure are more uniformly distributed along vertical axis, in prone position. It has been shown that in prone position in healthy subjects, regional lung volume at functional residual capacity was not influenced by gravity, in either breathing spontaneously or passively by mechanical ventilation. Rehder K, Knopp TJ, Sessler AD. Regional intrapulmonary gas distribution in awake and anesthetized-paralyzed prone man. J Appl Physiol 1978;45(4):528-535.

Therefore, in prone position alveolar stress-strain is more homogeneously distributed, which assumes very important role in terms of VILI, in ARDS.

Because of the inherent anatomy, rib cage is more compliant ventrally than dorsally. In prone position, ventral chest wall movement becomes relatively restricted, depending upon the type of padding, resulting in reduced chest compliance.

Effect of prone position on respiratory system compliance, depends upon change in lung compliance. Several observational studies have shown that respiratory system compliance remained unaltered or modestly decreased in prone position in surgical subjects as well as ARDS patients⁸. Change in lung compliance is determined by the degree of recruitment achieved in prone position. If improvement in lung compliance exceeds reduction in chest wall compliance, respiratory system compliance increases.

Therefore, regional ventilation is nearly homogenous in prone position, in both spontaneous and positive pressure ventilation. This results in more uniform stress (pressure per unit volume) and strain (deformability of neighboring alveoli) distribution.

As pulmonary perfusion remains unaffected with position change and preferentially distributed towards dorsal regions, regional V/Q is more homogenously matched in prone position. It has been demonstrated in experimental studies that prone position improves V/Q matching. In healthy dogs and pigs, V/Q matching improved from 0.83 to 0.94 and 0.72 to 0.82 respectively, from supine to prone position^9,10.

Thus, prone position results in favorable lung mechanics (stress-strain) and gas exchange.

 

PRONE POSITION AND CARDIOVASCULAR PHYSIOLOGY

Right ventricular dysfunction and acute cor pulmonale is not an uncommon complication of ARDS and mechanical ventilation. Incidence of acute cor pulmonale is around 22-25% in ARDS, which increases to 50% in severe ARDS with PaO₂:FiO₂ ration less than 100 ¹¹. Acute cor pulmonale increases mortality in ARDS from 36% to 60%¹².

Respiratory acidosis further worsens right ventricular function. Application of higher PEEP requires higher driving pressure (P_PLAT-PEEP) to optimize alveolar ventilation to avoid worsening of permissive hypercapnia and consequent acute cor pulmonale. Prone position achieves the same goals with less PEEP, minimizing these complications. Therefore, prone position in moderate to severe ARDS is essentially a right heart protective ventilation.

Prone position ensures lung protection (stress-strain), heart protective (right ventricular function) while improving gas exchange.

It appears that pulmonary mechanics is developed and optimized, evolutionarily for prone position, which is the normal posture in quadrupeds. 

Favorable changes in pulmonary mechanics in prone position becomes more pronounce in ARDS lung, mainly in terms of stress-strain and iotrogenic lung injury.

In ARDS increased weight of lung superseded the other factors (shape matching, heart weight, abdominal pressure) in causing compression of dependent lung regions in supine position. In other words, the decremental vertical gradient in the size of alveoli becomes exaggerated. The baby lung (with normal compliance) in non-dependent regions is hyperinflated and bears the brunt of stress and strain. While collapse alveolar units at dependent region remain closed throughout the respiratory cycle.

Vertical gradient in pleural pressure increases significantly in supine position (0.53 to 0.71 cmH₂O), compared to prone position (from 017 to 0.27 cmH₂O) in response to hypervolemia and increased extravascular lung water¹³. Hypervolemia induces intraabdominal hypertension, which decreases chest wall compliance and marked increases pleural pressure in dorsocaudal regions of lung (from -0.02 to +4.2 cmH₂O)¹⁴. Therefore, fluid resuscitation in sepsis or trauma induced ARDS, aggravates dependent airway closure and alveolar heterogeneity, more markedly in supine position than in prone position.

Prone position improves oxygenation by decreasing V/Q heterogeneity, which is more salient in presence of intraabdominal hypertension ¹⁵.

Response of PEEP on pulmonary perfusion is different in supine and prone position. In supine position, high PEEP exaggerates gravitational influence on perfusion distribution. Consequent redistributing of pulmonary perfusion, away from upper and middle zones to lower zones of lung, aggravates V/Q heterogeneity, worsening gas exchange¹⁶. In prone position, gravitational effect on pulmonary perfusion remains unaffected, by PEEP. In fact, decreased perfusion heterogeneity with PEEP has been found in prone position¹⁷. 

This further reinforces that gravity has little effect on pulmonary perfusion, which is determined mainly by anatomic and physiologic factors in pulmonary vasculature.

PEEP results in more uniform transpulmonary pressure and alveolar volume distribution in prone position without affecting pulmonary perfusion distribution. This leads to more homogenous stress-strain distribution and V/Q matching.

Prone position in ARDS decrease the large inequality in sizes of alveolar units, resulting in less heterogeneity. PEEP lead to opening of dependent closed alveolar units (recruitment) further reducing the heterogeneity. Previously atelectatic alveolar units come to be in non-dependent position while hyperinflated baby lung become dependent. Thus, the hyperinflation of baby lung is reduced, whereas previously closed lung units, open and participate in ventilation (improved compliance and improved ventilation. This is translated into improved V/Q matching, decreased regional stress and strain and thus decreased propensity to VILI. Using quantitative CT scan in ARDS patients has shown that prone position enhances the efficiency of PEEP by improving recruitment while reducing alveolar hyperinflation.

The compliance of lung in prone positon in the consequence of amount of opened alveolar units (recruitable lung tissue). If lung compliance improves more than the fall in chest wall compliance, respiratory system compliance increases. This is reflected in reduced distending pressure in mechanical ventilation. With increasing severity of ARDS, recruitable lung tissue increases. This may be reason of improved mortality benefit of proning in patients with PaO2:FiO2 ratio less than 100. Thus prone position confers its benefit in ARDS patients who meet the preconditions for prone position to work. Recruitability of lung tissue decreases over time because of structural changes in the lung tissues. Thus, precondition for prone position is met in severe ARDS, early in the disease onset.

PRONE POSITION AND GAS EXCHANGE:

To improve oxygenation, alveoli are required to remain inflated only (apneic oxygenation). They need not be ventilated. For CO2 clearance, alveoli are required to participate in ventilation also. Thus, response of prone positon to oxygenation and ventilation (CO2 clearance) may follow different pattern, which has its effect on mortality¹⁸. Oxygenation increases mainly due to improved V/Q homogeneity in both vertical and horizontal plane, which is consequence more uniform transpulmonary pressure and alveolar volume distribution.

Response to improved oxygenation is seen when prone position is initiated early (within 3 days), during exudative phase where congestive and atelectasis are predominant ^19,20. In later phase of ARDS, fibrosis prevails, which does not allow prone positon related improvement in pulmonary mechanics.

Improvement in oxygenation follow as path of rapid increase (within 30 minutes), followed by gradual rise over an extended and variable period, regardless of etiopathogenesis of ARDS⁸.

Improvement in ventilation (CO2 clearance) is a function recruitment of non-aerated dependent alveoli, and reduced hyperinflation of non-dependent alveoli. Subsequent improvement in lung compliance, which exceeds the decline in chest wall compliance, results in improvement in alveolar ventilation even without improvement in minute ventilation (decreased dead space ventilation) and decreased stress and strain. Reduction in CO₂ is better indicator of lung recruitment in ARDS, as it is twenty time more diffusible than O₂ and may predict reduced incidence of VILI and mortality²¹.

 

PRONE POSITION AND VILI:

Prone position delays the incidence of VILI by 50-80% ^22,23. In prone position, more homogenous distribution of alveolar volume, both vertical and horizontal plane, results from uniform pleural pressure and transpulmonary pressure distribution. This leads to decreased stress riser and more uniform stress-strain distribution throughout the lung. Improved alveolar homogeneity means higher static strain and lesser dynamic strain, for a given end inspiratory lung volume. Prone position also reduced the releases of inflammatory mediators ^24,.

Therefore, prone position reduces barotrauma, volutrauma, atelecto-tauama as well as biotrauma.

Prone position relieves the compression of heart on supporting lung, alleviating decremental gravitational gradient on alveolar sizes. It also improves lymphatic drainage, thereby reducing hydrostatic lung edema.

Prone position facilitates postural drainage of airway secretions, thereby reducing airway resistance, risk of bacterial contamination /infection and surfactant depletion.

Prone position in ARDS, by improving oxygenation and ventilation, reduces the need for frequent interventions and thus iatrogenic complications. It also gives room to optimize intravascular volume, by decreasing the requirement of more negative fluid balance and consequent vasopressor use.