| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Eur J Cardiothorac Surg 2004;26:554-563
© 2004 Elsevier Science NL
a Research Center, Montreal Heart Institute, 5000 Belanger Street, Montreal, Que., Canada H1T 1C8
b Department of Surgery, Montreal Heart Institute, 5000 Belanger Street, Montreal, Que., Canada H1T 1C8
c Department of Pharmacology, University of Montreal, C.P. 6128, succ. Centre-ville, Montreal, Que., Canada H3C 3J7
d Department of Anesthesiology, Notre-Dame Pavilion (CHUM), 1560 Sherbrooke East, Montreal, Que., Canada H2L 4M1
Received 3 December 2003; received in revised form 15 April 2004; accepted 5 May 2004.
* Corresponding author. Address: Research Center, Montreal Heart Institute, 5000 Belanger Street, Montreal, Que., Canada H1T 1C8. Tel.: +1-514-376-3330x3471; fax: +1-514-376-1355
e-mail: lpperrau{at}icm.umontreal.ca
 |
Abstract |
|---|
 Top Abstract 1. Introduction2. Material and methods3. Results4. Discussion5. ConclusionReferences |
|---|
Key Words: Endothelium • Pulmonary arteries • Cardiopulmonary bypass • Ventilation • NO inhalation
|
1. Introduction |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. Discussion5. ConclusionReferences |
|---|
Since prolonged intubation is associated with higher morbidity and mortality following cardiac surgery, several methods have been attempted to limit the respiratory dysfunction following CPB [4]. Pulmonary ventilation during CPB remains controversial since it involves a moving surgical field (which, however, also occurs during off-pump coronary artery bypass (OPCAB) surgery) and it may lead to certain adverse effects, as bronchoconstriction secondary to alveolar hypocapnia, which was reported by Bayindir et al. [5]. Nevertheless, other authors have reported better oxygenation after CPB using a vital capacity maneuver during weaning from bypass [6]. Nyhan and colleagues [7] have shown in a chronic canine model, that exposure to CPB for 150 min induces a selective endothelial dysfunction to acetylcholine (ACh) documented 4 days after CPB which is normalized by 14 days postoperatively [7]. The present study focuses on documenting the physiological consequences of CPB on pulmonary vessels in the short term/acute phase since clinically relevant pulmonary complications occur mainly in the hours following surgery. The use of ventilation during CPB being unexplored in controlled trials, this study will compare different strategies to limit pulmonary endothelial dysfunction and postoperative hypoxia after beating heart CPB.
|
2. Material and methods |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. Discussion5. ConclusionReferences |
|---|
2.2. Experimental groups
Group 1: Control (n=6). After induction of general anesthesia, a median sternotomy was performed, the animal was exsanguinated, and the heart and lungs were harvested en bloc.
Group 2: Sham without CPB (n=6). With the piglets in a supine position, EKG probes and the rectal thermometer probe were installed. Then, the swine was shaved, disinfected and drapped with sterile fields. The carotid artery was cannulated to monitor the systemic arterial pressure. The femoral vessels were isolated on both sides and a median sternotomy was performed. The pericardium was suspended using silk 5-0 and a double purse string was made on the right atrial appendage using a prolene 4-0.
Three minutes after intravenous heparin administration (3 mg/kg), a blood sample was drawn from the right atrium and the level of anticoagulation was assessed using an activated coagulation time (ACT) with Hemochron 801 (Technidyne, NJ). When the ACT was superior to 200 s, the atrial appendage and the right femoral artery were cannulated with a dlp 30-Fr double-staged (Medtronics, Mississauga, ON) and a Bardic 12-Fr (USCI, Division of BARD, NY), respectively. After 150 min, the heart and lungs were harvested en bloc.
Group 3: CPB 150 min no reperfusion (n=4). The same procedure as in group 2 was followed with the use of CPB. The jugular vein was cannulated and a bolus of fentanyl (Abbott Laboratories, Ville St-Laurent, QC) 15 µg/kg was given in 10 min. Anesthesia was maintained using a continuous infusion of fentanyl (2500 µg) and midazolam (Sabex, Boucherville, QC) (100 mg) given at a rate of 15 ml/h. The carotid artery was cannulated to obtain the arterial pressure. A cystostomy was performed to monitor the urine output per CPB.
CPB was initiated when the ACT was superior to 400 s. The bypass circuit consisted of a filtered hardshell venous reservoir (Minimax 1316, Medtronics, Mississauga, ON), a hollow fiber membrane oxygenator (Minimax Plus 3381 without Carmeda Bioactive Surface, Medtronics, Mississauga, ON), a heater–cooler, and a roller pump (Sarns 7000, Ann Arbor, MI). No arterial filter was used. The circuit was primed with 15 mequiv. of bicarbonate, 25 ml of mannitol, 5000 units of heparin, 300 ml of lactated ringer and 300 ml of pentaspan. After initial stabilization, the pump flow was adjusted to obtain an index of 2.4 l/min per m2 and the mean systemic arterial pressure was maintained between 50 and 75 mm Hg using phenylephrine (PE) as needed. After obtention of a steady state under CPB, mechanical ventilation was stopped and the body temperature of the animal was allowed to drift to 34 °C. The heart was left beating during CPB and no aortic cross-clamping was applied. After 130 min of CPB, the animal was rewarmed to 37 °C. At 150 min of CPB the pump was stopped, the animal was immediately sacrificed, and the heart and lungs were harvested en bloc.
Group 4: CPB 150 min+reperfusion 60 min (n=6). The same procedure was followed as in group 3. After 150 min on pump, mechanical ventilation with isoflurane was reinstituted and the animal was weaned from CPB. PE was started at the completion of the CPB to maintain the hemodynamics within normal range when needed. After 60 min of pulmonary reperfusion (following CPB weaning), the heart and lungs were harvested en bloc.
Group 5: CPB 150 min+ventilation (tidal volume 12 ml/kg)+reperfusion 60 min (n=4). The same procedure was followed as in group 4, except mechanical ventilation was maintained during the whole experiment.
Group 6: CPB 150 min+ventilation (tidal volume 12 ml/kg)+NO inhalation (NO 40 ppm)+60 min of reperfusion (n=4). The same procedure as in group 4 was followed with continuous mechanical ventilation during the whole experiment and inhalation of NO 40 ppm for the duration of the 150 min CPB run.
NO was administered using a NO apparatus developed by the Department of Anesthesiology in conjunction with the Department of Biomedical Physical Physics at the Notre-Dame Pavilion (CHUM, QC). This apparatus was calibrated prior to experiment and injected NO cyclically with a precision flowmeter during the inspiratory phase. NO was measured using a NO/NO2 electrode.
2.3. Vascular reactivity studies
After harvesting, the heart and lungs were rapidly placed in a modified Krebs-bicarbonate solution. Oxygenation was ensured using a carbogen mixture (95% O2 and 5% CO2). The heart was removed and branches of second-degree pulmonary arteries were isolated and dissected free of connective tissue and adventitial tissue and cut into 4–5 mm long rings. A total of 16 pulmonary arterial rings were prepared from each animal.
The vascular reactivity of rings of pulmonary arteries was studied in organ chambers filled with modified Krebs-bicarbonate solution (20 ml at 37 °C), and oxygenated with a 95% O2 and 5% CO2 mixture. The rings were suspended between two metal stirrups, one being connected to an isometric force transducer. Data were collected with a data acquisition software (IOS3, Emka Inc., Paris, France).
Each preparation was stretched to its active length–tension curve (usually 3.5 g), as determined by measuring the contraction to potassium chloride (30 mM) at different levels of stretch, and then stabilized for 30 min. The maximal contraction was determined with potassium chloride (60 mM) and rings were excluded if they failed to contract to potassium chloride (exclusion rate <5%). The baths were then washed and indomethacin (10–5 M) was added to exclude the production of endogenous prostanoids. After 60 min of stabilization, PE (range 2x10–7 to 3x10–6 M) was added to achieve a target contraction averaging 50% of the maximal contraction to KCl (60 mM).
2.3.1. Endothelium-dependent relaxations
The NO-mediated relaxation pathway was studied by constructing concentration–response curve to ACh (10–9 to 10–3 M), an agonist that binds to M2 receptors coupled to Gi-proteins and bradykinin (BK) (10–12 to 10–6 M) an agonist that binds to B2 receptors coupled to Gq proteins leading to the release of NO and the endothelium-derived hyperpolarizing factor.
Endothelium-independent relaxations were studied with sodium nitroprusside (SNP) (10–10 to 10–5 M; an exogenous NO donor) to study the integrity of smooth muscle cells.
2.4. Hemodynamic and biochemical data
Heart rate was continuously recorded from five subcutaneous limb electrodes. Arterial and venous blood gases were obtained from left and right atrium at regular intervals during the experiment (baseline, during CPB at 15, 45, 75 and 115 min, immediately after weaning from CPB, 30 and 60 min after weaning) and maintained within physiological limits by adjusting the ventilation rate and tidal volume. Pulmonary artery pressures were measured with a Swan-Ganz catheter at different intervals of the procedure: after induction and after weaning of CPB (30 and 60 min).
2.5. Cyclic GMP measurement
Basal cGMP level present in the second-degree pulmonary arteries of all groups was measured. Segments were frozen in liquid nitrogen and stored at –70 °C until the measurement of cGMP. The samples were subsequently pulverized and resuspended in trichloroacetic solution (TCA; 6.25%, w/v) to precipitate the proteins of the tissue. After centrifugation, the supernatant was washed with diethylether to preserve the cGMP and eliminate the TCA. Finally, the samples were heat dried by nitrogen gas to obtain purified cGMP. The cGMP was measured using enzymeimmunoassay system with acetylation based on rabbit anti-cGMP antibody (Amersham Pharmacia Biotech, Baie d'Urfé, QC). The levels were adjusted to the quantity of proteins measured in the tissue using the Bradford microassay technique (Bio-Rad, Mississauga, ON).
2.6. Endothelial coverage studies using silver nitrate staining
Silver nitrate segments of pulmonary arteries after 150 min+reperfusion 60 min was performed to assess the endothelial coverage. The rings were fixed first for 10 min in a paraformaldehyde buffer (4%). They were then washed for 1 min with Hepes sucrose buffer solution. Silver nitrate 0.25% was applied for 1 min. Washing was performed for 1 min before a second fixation for 2 min. The rings were exposed to ultraviolet light for 2–4 h in a cacodylate buffer solution. Preparation was read by a blinded investigator and representative photomicrographs were taken.
2.7. Drugs
All solutions were prepared daily. Potassium chloride, PE, indomethacin, ACh, BK, and SNP were purchased from Sigma Chemical Co., Oakville, ON.
2.8. Statistical analysis
Relaxation and contraction are expressed as a percentage of the maximal contraction to PE for the ACh group and expressed as mean±SEM; 16 rings of pulmonary artery were used per animal, n refers to the number of animals studied, the number of samples studied is nx16. ANOVA studies were performed to compare concentration–response curves. The Newman–Keuls test was used as the post hoc test. Student's test paired/unpaired observations were used for the comparison of the basal production of cGMP. Arterial pressure, pulmonary artery pressure, cardiac output, lactate levels and arterial oxygen pressure were compared using repeated measures ANOVA. Differences were considered to be statistically significant when the value of P was <0.05.
|
3. Results |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. Discussion5. ConclusionReferences |
|---|
|
|
|
|
|
3.2.2. Endothelium-dependent relaxations
There was a statistically significant decrease in endothelium-dependent relaxations to ACh in pulmonary rings in CPB with reperfusion compared with control, sham and CPB without reperfusion (Fig. 2a) . CPB with reperfusion induced a statistically significant decrease in endothelium-dependent relaxations to ACh compared to control, ventilation and ventilation with NO groups 1, 5 and 6 (Fig. 2b).
|
|
3.2.4. Cyclic GMP measurement
There was a statistically significant decrease of cGMP levels in all CPB groups when compared to control and sham. (Fig. 4)
.
|
|
4. Discussion |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. Discussion5. ConclusionReferences |
|---|
4.1. Vascular reactivity
4.1.1. Contraction
The decrease in endothelium-dependent and independent contractions to KCl and PE in CPB groups may be due to the expression of inducible nitric oxide synthase (iNOS) in the pulmonary vessels, creating a baseline vasodilatation.
4.1.2. Effect of reperfusion on vascular reactivity
During total CPB, as used clinically, pulmonary blood flow decreases greatly and the lungs are principally perfused by bronchial flow. This state of reduced pulmonary perfusion causes ischemia reperfusion injury to the lungs [8] which induces a significant decrease in endothelium-dependent relaxations to ACh suggesting reperfusion as the cause of endothelial dysfunction. This effect was observed in the present study as CPB without reperfusion did not decrease relaxations to ACh.
In a sheep model, total CPB for 90 min followed by 60 min of pulmonary reperfusion markedly altered pulmonary microvascular responses to ACh and serotonin [8]. Serraf and colleagues also observed a decrease in endothelium-dependent relaxations to ACh in pulmonary artery rings of piglets undergoing 90 min of CPB followed by 60 min of lungs reperfusion when compared to shams [9]. Nyhan and colleagues showed in a chronic canine model, that exposure to 150 min of CPB induced a selective endothelial dysfunction observed 4 days after CPB which normalized by 14 days [7] while relaxations to BK was not impaired.
Reperfusion after CPB is followed by a massive production of oxygen free radicals [10] which are generated by pulmonary endothelial cells during reoxygenation after hypoxia [11,12]. Oxidative injury may cause a functional uncoupling of the receptor/G-protein complex specific to the NO signal transduction pathway [17]. Reperfusion of the pulmonary tree significantly impairs relaxations to ACh mediated by Gi-protein [13]. The same observations were noted in this model, which shows alteration of endothelium-dependent relaxations to ACh and preserved relaxations to BK after CPB and reperfusion. The selective impairment of relaxations to ACh is not due to destruction or loss of endothelial cells since the relaxations to BK are preserved and the endothelial surface is intact as determined by histological examination with silver nitrate.
4.1.3. Effect of ventilation on vascular reactivity
Maintenance of mechanical ventilation during CPB prevented the occurrence of the impairment of endothelium-dependent relaxations in the present model. Becker and Sylvester showed in a ferret lung model that pulmonary ischemia did not cause hypoxia if ventilation was maintained while pulmonary blood flow was maintained at low levels. Furthermore, maintenance of pulmonary arterial pressures at physiological levels during ventilated ischemia attenuated lung injury by maintaining basal levels of NO production preserving rhythmic movement of fluid between alveolar and extraalveolar vessels generating shear forces and increasing intravascular volume thus diluting the toxic mediators accumulating during the period of ischemia [14,15]. The protective effect of ventilation during CPB may be due to preservation of the blood flow through the pulmonary artery.
However, Serraf and colleagues observed in a neonatal piglet model of CPB ventilated with 40 breaths/min, tidal volume 15 ml/kg an impairment of pulmonary artery relaxations to ACh [16] which may be due to a specific response of the neonatal pulmonary endothelium or to barotrauma or volutrauma.
4.2. Absence of additional effect of NO inhalation
The efficacy of NO inhalation therapy for postoperative pulmonary hypertension as a supplementation of endogenous endothelium-dependent relaxation factors has been established [17]. However, in a neonatal piglet model, NO ventilation (30 ppm) after CPB caused an impairment of pulmonary artery relaxations to ACh compared to controls [16]. These contradictions may be explained by the fact that NO inhalation started at the completion of CPB in Serraf's study. As oxygen becomes available at the start of reperfusion, xanthine oxidase creates a burst of superoxide anion production in tissue [3,18]. Supplemental NO may be chelated at that moment and increase free radical injury.
In the present study, NO inhalation during CPB preserved the advantages of ventilation, but did not give any additional benefit. The potential explanations are the dosage (only one dose was used) and a possible negative feedback on the NO pathway, limiting the endothelium-dependent relaxations in vitro after reperfusion. NO inhalation could possibly give other advantages on clinical endpoints, notably on oxygenation (not tested in this study).
4.2.1. Endothelium-independent relaxations
The lack of significant differences in endothelium-independent relaxations to exogenous NO donor SNP rules out an alteration of vascular smooth muscle cells as the cause for the endothelial dysfunction which is attributed to functional alterations of the signaling transduction mechanisms of endothelial cells.
4.2.2. Increased arterial oxygen pressure in the ventilated group
Since the amount of oxygen delivered to the animals during the experiments was maintained constant the same (FiO2=0.66), the greater partial arterial oxygen pressure in the ventilation group implies better pulmonary oxygen exchanges. The ratio of the arterial oxygen pressure over the fraction of inhaled oxygen (PaO2/FiO2) is frequently used in intensive care units as one of the criteria for Acute Respiratory Distress Syndrome (ARDS). A ratio below 300 is considered a criteria for acute lung injury and ARDS will be considered if the PaO2/FiO2 is below 200. In our study, the non-ventilated animals had an important decrease in the PaO2 following separation from CPB, the PaO2/FiO2 being as low as 166. In the ventilation group, this ratio was never lower than 330. Thus, ventilation in the CPB animals prevented the development of lung injury that could reach the magnitude of ARDS.
4.2.3. Cyclic GMP measurement
In the present study, CPB without reperfusion (group 3), CPB with reperfusion (group 4), ventilation (group 5) and NO inhalation (group 6) showed statistically significant decreases in the levels of cGMP when compared to the group 1: control. cGMP is formed by the enzyme guanylate cyclase after activation by NO, it thus represents an indicator of NO bioavailability. The decrease of lung tissue concentration of cGMP to 
70% of pre-CPB value after CPB has been previously reported [19]. The decrease in nitrite and nitrate post-CPB is independent of changes in eNOS, iNOS activity of gene expression [19]. In addition, Western blot analysis detected no changes in iNOS protein levels [19]. A decrease in NOS substrate or cofactor availability after CPB are possible mechanisms which warrant further studies [19]. From all this data, yet another mechanism could be inferred, CPB causes neutrophil activation leading to the formation of free radicals which could bind NO forming peroxynitrites thus causing a decrease in the bioavailability of NO which in turn could lead to a decrease in cGMP.
However, the reason for the decrease in cGMP in shams versus the controls is unclear. The activation of the alternate complement pathway in sham submitted to surgical trauma without CPB could possibly explain the difference in cGMP content between the control and sham.
Since CPB was not used in the sham, the statistically significant decrease in levels of cGMP in the CPB without reperfusion (group 3), CPB with reperfusion (group 4), ventilation (group 5) and NO inhalation (group 6) compared to the sham are inferred to the effect of CPB on the NO-cGMP axis.
4.3. Clinical relevance
Pulmonary endothelial injury, pulmonary hypertension and postoperative hypoxia contribute to mortality and morbidity in patients undergoing surgery with CPB. In the present study, ventilation during CPB prevented the occurrence of the endothelial dysfunction arising after reperfusion of the pulmonary arterial tree. It also prevented the development of lung injury of a magnitude of the ARDS. Mechanical ventilation during CPB could potentially benefit all patients undergoing cardiac surgery at no additional cost. With the advent of beating heart surgery or OPCAB, surgeons are becoming used to operate while the lungs are ventilated and should be less reluctant to use mechanical ventilation during CPB.
4.4. Limitation
The use of pulmonary arteries of young and healthy swine may not reflect adequately the adult patient population undergoing coronary artery bypass surgery, valvular surgery or transplantation which suffer from systemic hypertension, pulmonary hypertension, diabetes mellitus, dyslipidemia, heart failure and atherosclerosis associated with endothelial dysfunction. The function of G proteins was not assessed nor the role of vasoconstrictors such as endothelin or thromboxane A2 in the endothelial dysfunction. Aortic cross-clamping, cardioplegia and protamine were not used in this model contrary to the clinical practice [20] and could have intensified the observed pulmonary endothelial dysfunction. On the other hand, protamine, a polycationic protein rich in the aminoacide L-arginine, may cause pulmonary vasodilation [20]. The time course and reversibility of these pathological findings remain unknown as well as their relationship to the length of bypass time. The chosen ventilation strategy was based on standard intensive care practice, the use of different ventilation parameters could have led to similar outcomes.
|
5. Conclusion |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. Discussion5. ConclusionReferences |
|---|
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstract1. Introduction2. Material and methods3. Results4. Discussion5. ConclusionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  L.-M. Stevens, S. Fortier, M.-C. Aubin, I. El-Hamamsy, S. Maltais, M. Carrier, and L. P. Perrault Effect of tetrahydrobiopterin on selective endothelial dysfunction of epicardial porcine coronary arteries induced by cardiopulmonary bypass. Eur. J. Cardiothorac. Surg., September 1, 2006; 30(3): 464 - 471. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. A. Vohra, A. Levine, and J. Dunning Can ventilation while on cardiopulmonary bypass improve post-operative lung function for patients undergoing cardiac surgery? Interactive CardioVascular and Thoracic Surgery, October 1, 2005; 4(5): 442 - 446. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |
