|
|
||||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Division of Cardiothoracic Surgery, The Chinese University of Hong Kong, Prince of Wales Hospital, Hong Kong, China.
Correspondence: Ahmed A. Arifi, MD, Division of Cardiothoracic Surgery, The Chinese University of Hong Kong, Prince of Wales Hospital, Sha Tin, NT Hong Kong;
| Abstract |
|---|
|
|
|---|
Key Words: cardiac surgery cardiopulmonary bypass cytokine inflammatory response ischemia-reperfusion lung injury neutrophils ventilation
Postoperative pulmonary dysfunction in patients undergoing cardiopulmonary bypass (CPB) is a significant clinical problem and has long been recognized by cardiac surgeons, anesthetists, and intensive-care physicians. The disturbance may be manifested as conditions ranging from subclinical functional changes in most patients to full-blown ARDS in < 2% of cases after CPB.1 2 3 The mortality rate associated with ARDS is > 50%,1 2 not including the morbidity leading to prolonged postoperative recoveries and hospital stays. Despite years of research into this phenomenon, the understanding behind the complex pathophysiology of CPB- induced lung injury remains incomplete. We review the current knowledge on this subject, with particular emphasis on some therapeutic modifications.
| The Phenomenon |
|---|
|
|
|---|
Physiologic Changes
The physiologic disturbance after CPB can be categorized grossly into abnormal gas exchange and poor lung mechanics. The assessment of these functions has been measured through numerous parameters, such as the alveolar-arterial oxygen pressure difference (P[A-a]O2), intrapulmonary shunt function, the degree of pulmonary edema (ie, extravascular lung water), lung compliance, and pulmonary vascular resistance. Significantly increased P(A-a)O2 and pulmonary shunt fraction, together with decreased functional residual capacity and carbon monoxide transfer factor, have been observed in patients after CPB.4
5
Disturbances of lung mechanics in terms of poor static and dynamic lung compliance after CPB also were noted.6
7
In addition to these abnormalities, lung function tests after CPB in children and neonates have shown lower FVC levels, inspiratory capacities, and small airway flow rates.8
Lung disturbances after CPB include increased lung permeability9 10 11 12 13 14 and pulmonary vascular resistance,15 as well as lung surfactant changes. Pulmonary epithelial-capillary endothelial permeability is closely related to the formation of pulmonary edema, alveolar protein accumulation, and the facilitation of inflammatory cell sequestration, all of which consequently affect lung function. The increase in pulmonary permeability after CPB has been shown by the increased rate of transfer of Tc 99m-labeled diethylenetriamine pentaacetate,9 the protein accumulation index of radiolabeled transferrin,10 11 the systemic/bronchoalveolar urea ratio,11 the IV 67Ga level (transferring binder),12 and other similar indicators.13 As a result, the BAL sample protein content can increase by more than threefold to fourfold after a patient undergoes CPB.14 Finally, CPB could affect the pulmonary surfactant activity, particularly in infants and neonates.8 14 However, these changes appear to make no major contribution to the initial stages of lung injury after CPB.10
Biochemical Changes
Various biochemical changes can reflect the presence of lung injury after CPB. These include the substances directly or indirectly responsible for causing lung injury (eg, neutrophil elastase), or the products released from injured lung tissue (eg, 7S protein fragment of collagen or procalcitonin), and the reduction of products normally released by the lung (eg, nitric oxide [NO]).
Neutrophil elastase, a proteolytic enzyme, has long been measured as a marker of pulmonary injury after CPB both in the systemic circulation and in the BAL fluid. Tonz et al16 detected a positive correlation between systemic elastase peak concentrations and both the postoperative respiratory index and intrapulmonary shunt. However, the results of other studies have suggested that neutrophil elastase may not be a consistent marker of lung injury since no correlation between elastase concentration and gas exchange17 or acute lung injury score18 have been found.
The products associated with the breakdown of type IV collagen (a main constituent of basement membrane), such as the 7S protein fragment of collagen, have been used to mark lung injury. Increased 7S protein levels have been shown to be associated with high matrix metalloproteinase (MMP; proteolytic enzyme) and neutrophil concentrations in the BAL fluid of patients after they have undergone CPB.19 In addition, the lung is also a rich source of procalcitonin. The plasma concentration of procalcitonin can increase dramatically during pulmonary inflammation.20 Compared with many other inflammatory markers, a better correlation between high procalcitonin levels and the post-CPB Murray lung injury score has been observed.18
Decreased levels of exhaled NO were detected after CPB, citing a reduction of exhaled NO as a possible marker of pulmonary injury.21 22 It was proposed that the production of NO decreased after CPB because of transient pulmonary vascular endothelial or lung epithelial injury.21 Pearl et al22 correlated a reduced exhaled NO level to poor pulmonary compliance, elevated P(A-a)O2 levels, and elevated airway resistance after CPB. However, the exhaled NO level was found to have a poor correlation with plasma nitrite levels, indicating that decreased exhaled NO levels were mainly due to bronchial epithelial dysfunction rather than to pulmonary endothelial damage.22
Histologic Changes
Alveolar edema, extravasation of erythrocytes and neutrophils, and congested alveolar capillaries following CPB also have been confirmed by testing of intraoperative lung biopsy specimens.23
On electron microscopy, pneumocytes and endothelial cells appeared swollen and necrotic. Similar lung structural damage and changes after CPB also were observed on electron microscopy in animal models.24
| How Much of the Lung Injury Is CPB-Related? |
|---|
|
|
|---|
Lung Dysfunction After Major Surgery and CPB
It has been noticed that lung functional impairment is inevitable after any major surgery, a condition that most likely is related to the general anesthesia. Using CT scanning, researchers have found that general anesthesia induces atelectasis in nearly all patients.27
However, CPB appears to cause additional lung injury and to delay pulmonary recovery compared with other types of major surgery,5
conditions that generally are believed to be due to the damaging effects of a systemic inflammatory response associated with CPB.26
Yet, it is also noteworthy that the continuing refinement of CPB materials (ie, the use of a membrane oxygenator instead of a bubble oxygenator) as well as an improvement in anesthetic management (ie, early extubation leading to fast-track recovery) have largely limited such an additional lung injury.17
Hypothermic vs Normothermic CPB
The impact of temperature during CPB on lung function has been controversial. Birdi et al28
found that the perfusion temperature did not significantly influence gas exchange (P[A-a]O2) after coronary artery bypass grafting (CABG). However, reduced values for intrapulmonary shunt function, P(A-a)O2, and alveolar-arterial CO2 gradient were reported in the normothermic group in another study,29
indicating that normothermia may preserve lung function after CPB.
On-Pump vs Off-Pump CABG
With the re-emergence of off-pump CABG, interest has grown in the isolated effect of CPB on postoperative pulmonary dysfunction. Off-pump CABG was associated with reduced cytokine response when compared with on-pump CABG,30
31
32
and the attenuated inflammatory reaction may lead to less impaired postoperative lung function.
It has been shown that the number of circulating neutrophils and monocytes, as well as the level of neutrophil elastase, were significantly lower following off-pump CABG compared to on-pump CABG.32 33 Furthermore, oxidative stress, as indicated by the levels of lipid hydroperoxides and nitrotyrosines, was also significantly lower in the off-pump CABG group.33 Kilger et al34 detected a lower procalcitonin level, reflecting a reduced degree of lung injury, in patients undergoing off-pump CABG. Despite these findings, both on-pump and off-pump CABG patients experienced similar degrees of decreased PaO2 and increased P(A-a)O2, and higher percentage of pulmonary shunt fractions after undergoing CABG.6 17 35 As far as the duration of ventilatory support is concerned, the off-pump approach may be beneficial in high-risk patients undergoing repeat CABG,36 but not in those patients undergoing primary CABG.17 35 37 Therefore, although CPB is known to cause disturbances in lung mechanics,6 CPB itself may not be a major contributor to the postoperative gas exchange abnormalities.35
| Current Understanding |
|---|
|
|
|---|
,41
42
have been shown to promote PMN activation and recruitment. In addition, platelet-activating factor, leukemia inhibitory factor, and the arachidonic acid derivative leukotriene (LT) B4 also can contribute to this process.44
Nevertheless, TNF-
does not appear to be responsible for causing neutrophil sequestration.45
IL-6 also may play either an anti-inflammatory or a proinflammatory role in lung injury under different conditions.46
47 Activated PMNs can further release a number of proteolytic enzymes and oxidative chemicals both into the systemic circulation and into local lung tissue. These substances include degrading MMPs,48 PMN elastase,16 and oxygen-free radicals (ie, myeloperoxidase, hydrogen peroxide, and superoxides).38 These enzymes are instrumental in the development of post-CPB lung injury by breaking down the pulmonary ultrastructure, which results in increased pulmonary alveolar-endothelial permeability, thereby affecting gas exchange and lung mechanics.16 24 49
Neutrophil-Pulmonary Endothelial Adhesion
An increased expression of cell surface adhesion molecules (ie, CD1850
and CD11b51
) following the activation of PMNs during CPB enhances neutrophil-pulmonary endothelial adhesion.52
Subsequently, it leads to further PMN activation and to local pulmonary neutrophil recruitment and sequestration, which may result in the release of neutrophil proteolytic enzymes that cause direct lung damage.
Intercellular adhesion molecule-1, a ligand of CD18, has been seen to have increased expression on pulmonary endothelial cells after CPB and has been associated with greater pulmonary neutrophil accumulation.53 Meanwhile, complement C3 also plays an important role in inducing CD18 expression,54 while C5a was particularly involved in the expression of endothelial adhesion molecule P-selectin.55 In patients with a preexisting disease like diabetes mellitus, a higher CD11b expression was noted after CPB when compared with nondiabetic patients.56 This higher level of expression may increase the risk of post-CPB complications.56
Neutrophil Elastase
Peak systemic neutrophil elastase levels were observed at the end of CPB and have long been associated with postoperative pulmonary injury. Elastase is currently believed not only to be a marker of PMN activation,48
57
but also to be responsible for causing direct injury by its proteolytic activity on lung microvasculature and on endothelial cadherins.58
Elastase release can be augmented by some other mediators such as IL-6.59
Meanwhile, elastase also may serve as an activator of MMP-9.60
The antagonistic effect of MMP-9 on antiproteases (ie,
-1-protease inhibitor, which is an inhibitor of elastase activity) can lead to a positive-feedback relationship between elastase and MMP-9.
Free Radicals
Systemic48
and BAL fluid11
myeloperoxidase levels have been reported to be highest at the end of CPB, but the contribution of myeloperoxidase to lung injury is still controversial. It has been proposed61
that increased free radical activity represents a potential risk for ARDS after CPB. In fact, hyperoxic CPB is widely used in cardiac operations, and there is concern about whether oxygenation may induce oxygen-derived free radicals. It has been suggested that hyperoxic CPB, compared with normoxic CPB, increases oxygen free radical damage to the lung, as reflected by lower vital capacity and FEV1 levels.62
Arachidonic Acid Metabolites
Many arachidonic acid metabolites have potent vasoactive properties. Prostacyclin and prostaglandin E2 can cause pulmonary vasodilation, while LT-C4 and thromboxane B2 (TXB2) tend to cause vasoconstriction. The exact role of these mediators in pulmonary function after CPB is not entirely clear. Interestingly, the lung was found to be a major source of TXB2 following ischemia and reperfusion, and the correlation between raised TXB2 and impaired lung function has been shown in a sheep CPB model.63
Patients with more severe lung injury (Murray lung injury score, grade 2) after CPB had lower prostaglandin E2 concentrations and higher TXB2 plasma concentrations, but LT-B4 and LT-C4 levels remained unchanged.47
Hence, the imbalance between these arachidonic acid metabolites, rather than their individual effects, may be more critical in the development of post-CPB pulmonary edema. It has been suggested64
that increased pulmonary cyclooxygenase-2 expression after CPB may contribute to the development of such an imbalance.
| Therapeutic Interventions or Modifications |
|---|
|
|
|---|
,26
although there was little effect on complement activation.65
66
67
In addition, methylprednisolone therapy can inhibit neutrophil CD11b expression51
and neutrophil complement-induced chemotaxis,67
thereby decreasing neutrophil activation and post-CPB neutropenia.43
However, it failed to limit PMN elastase activity.65
In a porcine model,68
post-CPB lung function (ie, P[A-a]O2, pulmonary vascular resistance, and extracellular fluid accumulation) was better preserved after pretreatment with methylprednisolone. However, in a randomized clinical trial, patients who received methylprednisolone during a sternotomy or at the onset of CPB had similar or higher postoperative P(A-a)O2 levels and pulmonary shunt function, as well as longer intubation times compared with control subjects.7
69
Furthermore, methylprednisolone therapy was unable to prevent poor postoperative lung compliance.7
69
Aprotonin also has been shown to limit TNF-
and neutrophil elastase release and to complement activation as well as neutrophil CD11b up-regulation following CPB.26
IL-8 levels in the BAL fluid and pulmonary neutrophil sequestration after CPB was inhibited after the use of aprotinin.70
Aprotinin priming of the CPB circuit may result in reduced postoperative morbidity and length of ICU stay.26
The inflammatory response, and pulmonary dysfunction in particular, also was attenuated following the use of aprotinin in patients undergoing heart transplantation.26
Leukocyte Depletion
Leukocyte depletion during CPB may limit the postoperative inflammatory response, as measured by reduced IL-8 production,71
although its beneficial effects on post-CPB pulmonary function have been inconsistent. In some studies,71
72
leukocyte depletion did not significantly improve postoperative PaO2 levels and pulmonary hemodynamics. In other studies, better preserved lung function73
74
75
76
and less free radical generation77
have been associated with leukocyte-depleted CPB.
Modification of Artificial Circuit
Heparin-coated circuits are associated with reduced activation of leukocytes and the release of cytokines, resulting in less inflammatory reactions following CPB.78
79
80
Compared with conventional circuits, heparin coating may improve lung compliance81
and pulmonary vascular resistance81
and may reduce intrapulmonary shunting.82
However, such benefits can be transient and may not be clinically significant.83
In addition, a comparison between the hollow fiber and the flat sheet membrane oxygenators has suggested that a greater pressure drop (which correspond to shear stress) across the latter is associated with a more pronounced activation of leukocytes during clinical CPB and, therefore, should be avoided.
Continuous Hemofiltration
High-volume continuous hemofiltration, by potentially removing destructive and inflammatory substances from the circulation during CPB, can significantly reduce systemic edema, pulmonary hypertension, and improve lung function (eg, pulmonary vascular resistance, lung dynamic compliance, and P[A-a]O2) after CPB.84
Continuous hemofiltration also may reduce lung tissue malondialdehyde levels.84
Maintaining Mechanical Ventilation During CPB
Cardiac surgeons have accepted the common practice of stopping ventilation during CPB, as blood oxygenation by the lungs is no longer required and the movement from mechanical ventilation may interfere with the surgery. It is known that hypoventilation during CPB is associated with the development of microatelectasis, hydrostatic pulmonary edema, poor compliance, and a higher incidence of infection.85
86
Hence, some investigators63
85
86
87
have hypothesized that mechanical ventilation during CPB may limit postoperative lung injury by preventing these complications. Moreover, the lungs are totally dependent on oxygen supply from the bronchial arteries in the period of cardiac arrest. The additional contribution to lung tissue oxygenation via gas diffusion by continuous ventilation is therefore worth measuring, since a certain degree of pulmonary ischemia may exist during CPB when ventilation is stopped.
The effects of ventilation during CPB have been tested in a number of studies using vital capacity maneuver (VCM; ie, a peak airway pressure of 40 cm H2O with a fraction of inspired oxygen of 0.4 for about 15 s), continuous positive airway pressure (CPAP), and continuous ventilation over the period of cardiac arrest. In a porcine model, VCM at the end of CPB resulted not only in improved gas exchange,86 but also reduced the incidence of atelectasis, as determined by a CT scan soon after CPB.85 However, repeating the VCM may not produce extra benefits.86 Meanwhile, better postoperative gas exchange and less pulmonary shunting were observed in patients who received CPAP during CPB,87 although the beneficial effects of CPAP were not evident in an animal CPB model.88
To date, the evidence for the benefits of maintaining ventilation alone during CPB is inconsistent, with most studies showing no significant preservation of lung function. Continuous ventilation during CPB was shown to provide no significant improvement in pulmonary vascular resistance, cardiac index, or oxygen tensions in a pig model.89 Similarly, no differences in pulmonary epithelial permeability were found between ventilated and nonventilated patients undergoing CPB.90 However, maintaining ventilation together with pulmonary artery (PA) perfusion during CPB may be advantageous. Friedman et al63 compared total CPB (ie, no ventilation or PA perfusion) with partial CPB (ie, with ventilation and PA perfusion) in a sheep model. They suggested that ventilation with PA perfusion during CPB may have a beneficial role in preserving lung function by limiting platelet and neutrophil sequestration and attenuating the TXB2 response after CPB.63 More recently, it has been reported that liquid ventilation during CPB may increase oxygen delivery and may decrease pulmonary vascular resistance when compared with air ventilation in a neonatal porcine model.91 Furthermore, liquid ventilation at a higher functional residual capacity was more effective in optimizing postoperative alveolar distension and lung volume.91
Maintaining Lung Perfusion During CPB
Since the early days of open heart surgery, it has been recognized that CPB is associated with pulmonary ischemic-reperfusion injury. However, as far as preventing tissue ischemia during CPB is concerned, the lungs remain one of the least protected organs. The lungs have a bimodal blood supply from the PAs and the bronchial arteries, with extensive anastomotic connections. The bronchial arteries contribute about 1 to 3% of the total blood flow to the lungs under normal physiologic conditions. The relative contribution of the bronchial arteries and PAs, and of alveolar ventilation in delivering oxygen to maintain lung tissue viability is still unclear. The lungs are purely dependent on the bronchial arteries during CPB to provide the 5% of whole-body oxygen uptake that is necessary even under hypothermic condition.92
However, from the experience of lung transplantation, it is known that the bronchial arteries can be sacrificed without causing any obvious pulmonary dysfunction. In addition, ischemia and reperfusion during CPB can enhance the regional release of inflammatory mediators, which may further induce lung injury. For instance, it has been demonstrated93
94
that several proinflammatory cytokines, which are released from the heart during CPB, could be cleared in the lungs. Hence, it would be interesting to study whether maintaining PA perfusion during CPB can attenuate the deterioration of lung function.
A number of therapeutic strategies for lung ischemia-reperfusion injury have been investigated. Improved lung function was seen after CPB with PA perfusion compared with non-PA perfusion in animals,89 95 neonates,96 and adults,97 demonstrating the potential beneficial role for maintaining PA perfusion during CPB. In animal models, CPB without PA perfusion resulted in significantly higher pulmonary vascular resistance and P(A-a)O2 levels,89 with lower pulmonary compliance.95 Liu and colleagues98 also have shown that the use of a hypothermic anti-inflammatory solution for PA perfusion may help to prevent lung injury, as measured by better post-CPB pulmonary histology and lung function, as well as lower plasma malondialdehyde levels. In infants undergoing CPB, better-preserved lung function (ie, an increased PaO2/fraction of inspired oxygen ratio and shorter intubation time) was observed in a PA perfusion group compared with control subjects.96
Richter and colleagues97 have reported an attenuated cytokine response (ie, IL-6 and IL-8) and better-preserved lung function (ie, less pulmonary shunting, improved P[A-a]O2 levels and respiratory indexes, and earlier extubation) in patients undergoing bilateral extracorporeal circulation (ie, the Drew-Anderson technique). Whether the observed benefits resulted from maintaining lung ventilation and PA perfusion or from the avoidance of an extracorporeal oxygenator needs further clarification. Another recent study99 in a canine model demonstrated that the use of biventricular CPB helped in preserving lung function (ie, reduced pulmonary vascular resistance and extravascular lung water, with improved lung compliance) compared with conventional heart-lung bypass, which may represent further supportive evidence for maintaining PA perfusion during CPB.
| Summary |
|---|
|
|
|---|
Our current research interests focus on ventilation and PA perfusion during clinical CPB. Some recent encouraging results have shown that maintaining lung ventilation and PA perfusion during CPB potentially can minimize postoperative lung injury. Further elucidation of the underlying mechanism will help to refine therapeutic strategies for patients who need cardiac surgery.
| Footnotes |
|---|
This study was supported in part by the Direct Grant for Research (Chinese University of Hong Kong No. CRE-2001021) and the Research Grant Council Earmarked Grant (Chinese University of Hong Kong No. 4310/99M), Hong Kong SAR.
Received for publication March 20, 2001. Accepted for publication May 31, 2001.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
A. Y. Denault, Y. Lamarche, P. Couture, F. Haddad, J. Lambert, J.-C. Tardif, and L. P. Perrault Inhaled milrinone: a new alternative in cardiac surgery? Seminars in Cardiothoracic and Vascular Anesthesia, December 1, 2006; 10(4): 346 - 360. [Abstract] [PDF] |
||||
![]() |
E. H. J. Hulzebos, P. J. M. Helders, N. J. Favie, R. A. De Bie, A. Brutel de la Riviere, and N. L. U. Van Meeteren Preoperative intensive inspiratory muscle training to prevent postoperative pulmonary complications in high-risk patients undergoing CABG surgery: a randomized clinical trial. JAMA, October 18, 2006; 296(15): 1851 - 1857. [Abstract] [Full Text] [PDF] |
||||
![]() |
R K Kharbanda, J Li, I E Konstantinov, M M H Cheung, P A White, H Frndova, J Stokoe, P Cox, M Vogel, G Van Arsdell, et al. Remote ischaemic preconditioning protects against cardiopulmonary bypass-induced tissue injury: a preclinical study Heart, October 1, 2006; 92(10): 1506 - 1511. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. C Clark Lung Injury after Cardiopulmonary Bypass Perfusion, July 1, 2006; 21(4): 225 - 228. [Abstract] [PDF] |
||||
![]() |
A. Della Corte, M. Scardone, G. Romano, C. Amarelli, A. Biondi, L. S. De Santo, M. De Feo, G. Nappi, and M. Cotrufo Aortic Arch Surgery: Thoracoabdominal Perfusion During Antegrade Cerebral Perfusion May Reduce Postoperative Morbidity Ann. Thorac. Surg., April 1, 2006; 81(4): 1358 - 1364. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Perrin, A. Roch, P. Michelet, M. Reynaud-Gaubert, P. Thomas, C. Doddoli, and J.-P. Auffray Inhaled nitric oxide does not prevent pulmonary edema after lung transplantation measured by lung water content: a randomized clinical study. Chest, April 1, 2006; 129(4): 1024 - 1030. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Verheij, A. van Lingen, P. G. H. M. Raijmakers, E. R. Rijnsburger, D. P. Veerman, W. Wisselink, A. R. J. Girbes, and A. B. J. Groeneveld Effect of fluid loading with saline or colloids on pulmonary permeability, oedema and lung injury score after cardiac and major vascular surgery Br. J. Anaesth., January 1, 2006; 96(1): 21 - 30. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. M. Morariu, B. G. Loef, L. P. H. J. Aarts, G. W. Rietman, G. Rakhorst, W. van Oeveren, and A. H. Epema Dexamethasone: Benefit and Prejudice for Patients Undergoing On-Pump Coronary Artery Bypass Grafting: A Study on Myocardial, Pulmonary, Renal, Intestinal, and Hepatic Injury Chest, October 1, 2005; 128(4): 2677 - 2687. [Abstract] [Full Text] [PDF] |
||||
![]() |
Y. Lamarche, O. Malo, E. Thorin, A. Denault, M. Carrier, J. Roy, and L.P. Perrault Inhaled but not intravenous milrinone prevents pulmonary endothelial dysfunction after cardiopulmonary bypass J. Thorac. Cardiovasc. Surg., July 1, 2005; 130(1): 83 - 92. [Abstract] [Full Text] [PDF] |
||||
![]() |
U. Abdel-Rahman, F. Ozaslan, P. S. Risteski, S. M |