The lung and the brain: a dangerous cross-talk

Brain or lung injury or both are frequent causes of admission to intensive care units and are associated with high morbidity and mortality rates. Mechanical ventilation, which is commonly used in the management of these critically ill patients, can induce an inflammatory response, which may be involved in distal organ failure. Thus, there may be a complex crosstalk between the lungs and other organs, including the brain. Interestingly, survivors from acute lung injury/acute respiratory distress syndrome frequently have some cognitive deterioration at hospital discharge. Such neurologic dysfunction might be a secondary marker of injury and the neuroanatomical substrate for downstream impairment of other organs. Brainlung interactions have received little attention in the literature, but recent evidence suggests that both the lungs and brain can promote inflammation through common mediators. The present commentary discusses the main physiological issues related to brain-lung interactions.

probably by promoting a higher rate of pulmonary complications as well as by altering neurologic outcome [4]. Overall, the information in regard to the multiplepathway cross-talk between the brain and lungs is quite limited [5].
In the study by Quilez and colleagues [1], diff erent variables, including lung function, plasma, and lung levels of cytokines as well as c-fos gene, were measured. c-fos is a marker of neuronal activation and is correlated with an increase in functional and metabolic activity in the brain, involved in the phenomena of neuronal plasticity, expressed in response to a wide range of stimuli, and implicated in processes such as gene transcription, apoptosis, or proliferation [6]. Th e main results of the study by Quilez and colleagues can be summarized as follows: (a) independently of higher or lower V T , mechanical ventilation 'per se' (compared with spontaneous breathing) induced neutrophil infi ltration, increased lung damage, and was associated with a greater release of infl ammatory markers in both the lung and plasma as well as c-fos activation in central amygdala, hippocampus, paraventricular hypothalamic nuclei, and supraoptic nucleus; and (b) higher V T increased c-fos in the retrosplenial cortex and hypotalamus and increased tumor necrosis factor-alpha in the plasma. Th erefore, mechanical ventilation promoted brain activation and the intensity of the response was increased with higher V T , suggesting a cross-talk between the brain and lungs.
Diff erent mechanisms may explain the possible negative eff ects of mechanical ventilation on the specifi c anatomical areas of the brain, even in the absence of previous lung disease [7]: (a) release of local pulmonary infl ammatory mediators in the bloodstream with activation of brain response and (b) altered regional cerebral perfusion due to increased mean airway pressure, reduced lymphatic drainage, and activation of the autonomic system ( Figure 1).
Brain injury-associated pulmonary dysfunction has long been attributed only to a greatly increased sympathetic activity with pulmonary venoconstriction and higher capillary permeability, but a recent study reported the role of systemic infl ammatory response with pulmonary infi ltration of neutrophils, cytokine release, and endothelial dysfunction triggered by an initial sympathetic Abstract Brain or lung injury or both are frequent causes of admission to intensive care units and are associated with high morbidity and mortality rates. Mechanical ventilation, which is commonly used in the management of these critically ill patients, can induce an infl ammatory response, which may be involved in distal organ failure. Thus, there may be a complex crosstalk between the lungs and other organs, including the brain. Interestingly, survivors from acute lung injury/ acute respiratory distress syndrome frequently have some cognitive deterioration at hospital discharge. Such neurologic dysfunction might be a secondary marker of injury and the neuroanatomical substrate for downstream impairment of other organs. Brainlung interactions have received little attention in the literature, but recent evidence suggests that both the lungs and brain can promote infl ammation through common mediators. The present commentary discusses the main physiological issues related to brain-lung interactions.
discharge [8]. A clinical study has shown that, after severe brain injury, higher V T was associated with increased risk of acute lung injury (ALI) [9]. On the other hand, ALI itself induces a systemic infl ammatory response with elevated cytokines and neutrophils as well as a dysfunction of other organs, including the brain [10]. Protective ventilation could be associated not only with less VALI but also with better cerebral oxygenation and perfusion [11]. Heuer and colleagues [3] found that acute intracranial hypertension increased extravascular lung water and increased the amount of poorly aerated tissue in previously healthy lungs without alterations in gas exchange. In experi mental ALI, intracranial hypertension worsened pre-existing lung damage and further increased brain edema. Conversely, ALI alone increased the circulating concen trations of the neuronal damage markers (neuronal serum enolase and S100B), but the most severe hippo campal damage was seen in animals with combined ALI and intracranial hypertension. Th is is in agreement with clinical reports [11,12] and could indicate a reciprocal, synergistic eff ect of the two conditions.
Th e study by Quilez and colleagues, despite generating interesting data for the near future, has some limitations. First, small animals were used and thus similar results may not be obtained in larger animals with ALI. Second, rats were ventilated with very high V T (30 mL/kg, which is equivalent to 2,100 mL in average-sized adult humans) [13]. Th ird, protective mechanical ventilation was not applied since no positive end-expiratory pressure (PEEP) was used. Th erefore, the increase in brain damage and infl ammation, even with lower V T , could have been avoided by using moderate PEEP levels. Fourth, moderate hypercapnia and low pHa in the low V T group may have aff ected the results. Also, the duration of the experiment was quite short (3 hours), so that diff erent eff ects might be observed with longer periods of mechanical ventilation. Lastly, the type of sedation during spontaneous breathing and paralyzing agents during mechanical ventilation may aff ect lung infl ammation [14,15].
In conclusion, recent data from experimental and clinical studies have shown (a) increased evidence of a cross-talk between the brain and lungs and (b) a reduction in VALI as well as brain injury when protective ventilation is adopted. Th us, early detection of VALI, through clinical signs and measurements of extravascular lung water and of brain injury by measuring specifi c markers, could help to optimize ventilatory settings and even the overall clinical management of critically ill patients.

Competing interests
The authors declare that they have no competing interests.