| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ann Thorac Surg 2004;77:994-1000
© 2004 The Society of Thoracic Surgeons
a Department of Pediatric Cardiothoracic Surgery and Cincinnati, OH, USA
b Department of Cardiology, Cincinnati, OH, USA
c Department of Cincinnati Children's Hospital Medical Center; and Department of Surgery Cincinnati, OH, USA
d Department of Pediatrics, University of Cincinnati, Cincinnati, Ohio, USA
Accepted for publication September 8, 2003.
* Address reprint requests to Dr Pearl, Clinical Surgery, Pediatric Cardiothoracic Surgery, Cincinnati Children's Hospital Medical Center, 3333 Burnet Ave, OSB3, Cincinnati, OH 45229, USA.
e-mail: jeffrey.pearl{at}cchmc.org
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
METHODS: Crossbred piglets (5 to 7 kg) were cooled with cardiopulmonary bypass, followed by 120-min deep hypothermic circulatory arrest. Animals were then warmed to 38°C, removed from bypass, and maintained for 120 min. Methylprednisolone (60 mg/kg) was administered in the cardiopulmonary bypass pump prime (intraoperative glucocorticoids) or 6 hours before bypass (30 mg/kg) in addition to the intraoperative dose (30 mg/kg; preoperative and intraoperative glucocorticoids). Controls (no glucocorticoids) received saline.
RESULTS: Pulmonary vascular resistance in controls increased from a baseline of 152 ± 40 to 364 ± 29 dynes · s/cm5 at 2 hours of recovery (p < 0.001). Intraoperative glucocorticoids did not alleviate the increase in pulmonary vascular resistance (301 ± 55 dynes · s/cm5 at 2 hours of recovery, p < 0.001). However, animals receiving pre and intraoperative glucocorticoids had no increase in pulmonary vascular resistance (155 ± 54 dynes · s/cm5). Plasma endothelin-1 in controls increased from 1.3 ± 0.2 at baseline to 9.9 ± 2.0 pg/mL at 2 hours recovery (p < 0.01), whereas glucocorticoid-treated animals had lower endothelin-1 levels (4.5 ± 2.1 pg/ml, preoperative and intraoperative glucocorticoids; 4.9 ± 1.7 pg/mL, intraoperative glucocorticoids) at the end of recovery (p < 0.05). Intracellular adhesion molecule-1 in lung tissue was lower in animals receiving pre and intraoperative glucocorticoids (p < 0.05). Myeloperoxidase activity was elevated in control lungs at 2 hours of recovery compared with glucocorticoid-treated groups (p < 0.05). Inhibitor 
B
, the inhibitor of nuclear factor-B, was higher in lungs of animals receiving glucocorticoids compared with controls (p < 0.05).
CONCLUSIONS: Glucocorticoids prevented pulmonary hypertension after cardiopulmonary bypass and deep hypothermic circulatory arrest, which was associated with reduced plasma endothelin-1. Glucocorticoids also reduced pulmonary intercellular adhesion molecule-1 and myeloperoxidase activity. Inhibition of nuclear factor-B, along with reduced neutrophil activation, contributed to glucocorticoid alleviation of pulmonary hypertension after cardiopulmonary bypass and deep hypothermic circulatory arrest.
Abbreviations: CPB • cardiopulmonary bypass • DHCA • deep hypothermic circulatory arrest • FiO2 • fraction of inspired oxygen • GAPDH • glyceraldehyde-3-phosphate dehydrogenase • GC • glucocorticoids • ICAM-1 • intercellular adhesion molecule-1 • IB • inhibitor kappa B • LV • left ventricle • NF-B • nuclear factor-kappa B • NOS • nitric oxide synthase • PVR • pulmonary vascular resistance • PO2 • partial oxygen pressure • PCO2 • partial carbon dioxide pressure • RV • right ventricle
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
It is well established that CPB induces a systemic inflammatory response characterized in part by elevated proinflammatory cytokines, such as interleukins and tumor necrosis factor- (TNF-) [4–7]. Ischemia and reperfusion stimulate the potent pulmonary and coronary vasoconstrictor endothelin-1 and downregulate endothelial nitric oxide synthase (NOS) activity and endothelial nitric oxide production [2, 8–10]. Our laboratory previously demonstrated that reoxygenation following acute hypoxia resulted in significant pulmonary hypertension and impaired gas exchange that was associated with marked elevations in endothelin-1 and decreased nitric oxide levels [2].
Glucocorticoids are the mainstay of therapy aimed at decreasing the inflammatory response to CPB because of the known antiinflammatory properties, tolerance by patients, availability, and relative low cost [11–13] However, studies to date, have demonstrated that glucocorticoids do not completely suppress the CPB induction of proinflammatory cytokines, suggesting alternate mechanisms of action. Furthermore, the effectiveness and optimal timing of administration of steroids to reduce pulmonary injury in neonates and infants undergoing CPB have not been well characterized. Increasing evidence suggests that glucocorticoids might exert benefits by regulating transcription or translation of anti-inflammatory cytokines, such as interleukin-10 (IL-10), and altering expression of other proteins, such as endothelin-1 and inhibitor kappa B (IB) [14, 15]. Therefore, our group proposed that glucocorticoid administration has multiple mechanisms underlying improved cardiopulmonary recovery after CPB and DHCA in infants and children. This study focuses on mechanisms that improve pulmonary function in an immature piglet model and determines whether pre and intraoperative glucocorticoid administration could be more efficacious than intraoperative dosing alone.
|
Material and methods |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Piglets weighing 5 to 7 kg were anesthetized with ketamine (22 mg/kg, intramuscular) and acepromazine (1.1 mg/kg, intramuscular) then intubated, and mechanically ventilated. Continuous pentobarbital infusion (20 mg/kg per hour), intermittent fentanyl citrate (10 µg/kg per hour) and pancuronium bromide (0.1 mg/kg per hour) were used throughout the experimental protocol. The doses were sufficient to maintain deep general anesthesia. Median sternotomy exposed the heart and pressure catheters (Millar Instruments, Houston, TX) were placed in the pulmonary artery, right ventricle (RV), and left ventricle (LV). A Doppler flow probe (Transonics Systems, Inc, Ithaca, NY) placed around the pulmonary artery measured cardiac output. Baseline measurements of pulmonary artery and systemic arterial pressures, cardiac output, and respiratory function were taken after a 30-minute equilibration period. Hematocrit, electrolytes, PCO2, PO2, and pH were determined in arterial and mixed venous blood samples (Bayer Diagnostics, Emeryville, CA). A lung biopsy was obtained from the right lower lobe following baseline hemodynamic measurements.
Animals were administered heparin (300 U/kg) and placed on CPB with cannulation of the carotid artery and right atrium. The CPB pump prime consisted of 800 mL direct-drawn whole porcine blood (Animal Biotech Industries, Danboro, PA) stabilized with sodium citrate. A single dose of methylprednisolone (60 mg/kg) was administered intraoperatively in the pump prime in the intraoperative glucocorticoid group (n = 9), while a dose of 30 mg/kg was given six hours before CPB in addition to the intraoperative dose (30 mg/kg) in the pre and intraoperative glucocorticoid group (n = 9). Another group (no GC, n = 9) received equal volumes of saline but no glucocorticoids. Comparisons of tissue data were made between lung samples collected before and after CPB and DHCA. The steroid dosing and timing is similar to the clinical practice at our institution, which calls for either 30 mg/kg of methylprednisolone in the pump prime or, for neonates and preoperatively compromised infants, a dose of 30 mg/kg 4 to 6 hours before surgery and an additional 30 mg/kg in the pump prime [13]. The dosing and timing is based on the known pharmacokinetics and time to peak effect of methylprednisolone.
Once on CPB, animals were cooled to a rectal temperature of 18°C over more than 40 minutes. Hematocrit on CPB was maintained at 25% to 30% and calcium at 0.6 to 0.8 mmol/L with a flow rate of 100 mL/kg per minute. The PO2 of the bypass circuit was maintained at greater than 250 mm Hg. In treatment groups undergoing DHCA, the bypass circuit was then turned off and the head packed in ice. The heart was protected with topical cold saline and ice. Cold saline was also instilled into the pleural spaces. Circulatory arrest was maintained for 120 minutes. Cardiopulmonary bypass was reinstituted and the animals were warmed to 38°C over more than 45 minutes on CPB. Ultrafiltration during warming returned the hematocrit to 30% to 35%. Piglets were removed from CPB and maintained under anesthesia for 120 minutes. Dopamine at 5 µg/kg/min was administered during weaning from CPB then discontinued after 60 minutes. Cardiopulmonary function was monitored continuously during the period following CPB. Lungs were explanted immediately at terminal time points. Wet to dry tissue weight ratio was determined from the entire left lung and tissue analyses were performed on the right lung.
Hemodynamics were monitored throughout by recording ventricular pressures, systemic arterial pressure, pulmonary artery pressures (Ponemah Physiologic Platform; Gould Systems, Valley View, OH), and cardiac output (Transonic Systems, Inc.). Respiratory function is monitored by CO2SMO Plus respiratory profile system (Novametrix, Wallingford, CT). Dynamic compliance, airway resistance, and expired CO2 and O2 along with pulmonary artery pressures were monitored. Alveolar-arterial gradients, PO2:FiO2 ratio, and PVR were calculated for pulmonary function. Recovery values were compared with values recorded before CPB and DHCA for each animal.
Endothelin-1 measurements
Blood samples were immediately centrifuged at 4°C and plasma stored at -80°C for later analysis. A commercial human endothelin-1 immunoassay kit (R&D Systems, Minneapolis, MN) was used to measure protein concentration in plasma. The minimum detectable level of endothelin-1 is typically less than 1 pg/mL. The assay has less than 1% cross-reactivity with big endothelin.
Nitric oxide measurements
Exhaled nitric oxide (parts per billion) was measured with a chemiluminescence nitric oxide analyzer attached to the ventilator tube (Eco Physics, Inc, Ann Arbor, MI). The recorded level of exhaled nitric oxide was the mean of the peak measurements over three breathes.
Plasma levels of nitric oxide were estimated by measuring blood nitrate and nitrite using a two-part assay with nitrate converted to nitrite in the Griess reaction. The optical density at 550 nm, with correction at 650 nm, was measured. The data were analyzed by using GENESIS microplate software (Fisher Scientific Co, Pittsburgh, PA). Levels were reported as total nitrites (µmol/L). Although not a direct measure of systemic nitric oxide, nitrite/nitrate measurements have been established as an accepted methodology to estimate relative levels.
Measurement of mRNA expression
Ribonuclease protection assays determined levels of specific mRNA in lung tissue samples frozen in liquid nitrogen at the point of collection. A custom probe set containing DNA templates was used for T7 RNA polymerase-directed synthesis of antisense probes (PharMingen, San Diego, CA). The set included templates for probes that hybridize with porcine ICAM-1 and the housekeeping genes, L32 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The [32P] UTP-labeled probes were generated by in vitro transcription, then hybridized in excess to total mRNA (10 µg) isolated from porcine tissues. Free probe and single-stranded RNA were digested with RNAse A and T1. RNase-protected probes were purified by phenol-chloroform extraction, resolved on a 5% denaturing polyacrylamide gel, and imaged by autoradiography. The identity of each mRNA species in the original samples was determined based on the appropriate size of the protected probe fragment. Phosphorimaging of the gel allowed for the relative quantification of mRNA by Storm 860 software (Molecular Dynamics, Sunnyvale, CA). Housekeeping genes L-32 and GAPDH in the template allowed RNA levels to be normalized within gels. Messenger RNA expression is reported as the ratio of target mRNA to GAPDH to normalize for differences in background.
Tissue protein level analyses
Blood samples were centrifuged and plasma frozen at -80°C. Tissue samples were homogenized in 10 mmol/L 3-[N-morpholino] propane sulfonic acid buffer with protease inhibitors, centrifuged, and the supernatant frozen. Protein concentration was determined by the Bio-Rad protein assay and the samples are stored at -80°C until used. Western blots were performed with 30 µg total proteins separated on 4% to 12% acrylamide bis-tris gels (Invitrogen, Carlsbad, CA) by SDS-PAGE. Antibodies for immunoblotting were antiporcine ICAM-1, developed in our laboratory, and antihuman IB- (Santa Cruz Biotechnology, Santa Cruz, CA). Immunoblots were also probed with antibodies for GAPDH (0.05 µg/mL; Chemicon International, Temecula, CA). Secondary antibodies were alkaline phosphatase-conjugated goat antirabbit or antimouse IgG. Proteins were visualized with the Western Breeze chemiluminescent detection system according to the manufacturer's instructions (Invitrogen). Protein levels are reported as a ratio of target protein to GAPDH levels on the same immunoblot to correct for background effects.
Myeloperoxidase activity
The activity of myeloperoxidase, an enzyme occurring almost exclusively in neutrophils, is measured in lung tissue as a determinant of neutrophil infiltration. Frozen tissue samples (50 mg) are homogenized in 0.5% hexadecyltrimethylammonium bromide dissolved in 10 mmol/L 3-[N-morpholino] propane sulfonic acid, then centrifuged at 21,000g for 20 minutes at 4°C. The supernatant is mixed with sodium phosphate (80 mmol/L, pH 5.5) and tetramethyl benimide (16 mmol/L) and incubated at 25°C for 5 minutes. Hydrogen peroxide (1 mmol/L) is added and the samples incubated exactly 3 minutes at 25°C. A blank without hydrogen peroxide is also analyzed for each tissue. The reaction is stopped by the addition of 2 mol/L cold acetic acid. The optical density is measured at 650 nm on a spectrophotometer. Myeloperoxidase activity was the quantity of enzyme degrading 1 µmol hydrogen peroxide per minute at 37°C.
Statistical analysis of data
Repeated measures analysis of variance is used to analyze serial data over time and posthoc comparisons made by Fisher's PLSD test is used when appropriate to evaluate significant differences between individual time points. Comparisons between treatments are made by analysis of variance with a p value less than or equal to 0.05 considered significant. Personnel blinded to the treatment group status conduct analyses using Statview 4.01 software (Abacus Concepts Inc, Berkeley, CA).
|
Results |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
|
|
Nitrate and nitrite levels
Arterial nitrate and nitrite levels were reduced after CPB and DHCA in all treatment groups. Animals without glucocorticoids had reduced nitrate and nitrite levels 2 hours after CPB and DHCA. However, there was a smaller reduction in nitrate and nitrite levels in both groups receiving glucocorticoids (Table 2).
|
Lung myeloperoxidase activity
Myeloperoxidase activity was increased in all three groups after CPB and DHCA. However, myeloperoxidase activity at the end of recovery was lower in both groups receiving glucocorticoids compared with animals not receiving steroid treatment (p < 0.05; Table 2).
Pulmonary nuclear factor-B and inhibitor B
Although total nuclear factor-B (NF-B) levels in the lung tissue were not different among treatment groups, levels of the NF-B inhibitor, IB, were maintained in animals receiving glucocorticoids (p = 0.05). The ratios of IB to GAPDH protein at baseline were not different. The ratio 120-min after CPB and DHCA decreased without glucocorticoid administration but was maintained with intraoperative treatment and preoperative and intraoperative glucocorticoids (p < 0.05 vs no glucocorticoids, Table 2).
|
Comment |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
Endothelin-1 and nitric oxide
We have previously demonstrated that hypoxia and reoxygenation on CPB increases endothelin-1 levels and decreases exhaled nitric oxide and systemic nitric oxide levels in piglets [2]. Both hypoxia and reoxygenation stimulate endothelin-1 expression in vascular endothelial cells [18, 19]. The combination of decreased nitric oxide levels and increased endothelin-1 has been implicated as a major causative factor for postoperative pulmonary hypertension [8, 20–23]. Similarly, in the current study control animals exposed to CPB and DHCA exhibited significant elevations in PVR after CPB, which had a marked affect on hemodynamics.
Although the balance between endothelin-1 and nitric oxide levels is thought to be a prime regulator of PVR, endothelin-1 also appears to negatively impact nitric oxide synthase activity. The decrease in exhaled and systemic nitric oxide levels in conjunction with increased endothelin-1 has been reported to be prevented by endothelin-1 blockade [24]. Unlike endothelin-1 blockade, glucocorticoid therapy decreased, but did not prevent endothelin-1 effects on pulmonary endothelium.
The discrepancy between the intraoperative group and the combined pre and intraoperative group in regards to PVR, despite similar effects on endothelin-1 levels, may be a result of glucocorticoid-induced transcription and translation of other protective mechanisms. A plausible explanation for this, although not addressed in this study, is decreased nitric oxide levels related to ischemia-reperfusion induced decreases in nitric oxide synthase activity. Although nitric oxide synthase activity was not directly measured in this study, there was a trend toward higher systemic nitric oxide in preoperative and intraoperative treated animals, suggesting a beneficial effect of glucocorticoids on maintaining nitric oxide synthase activity. In fact, other studies have suggested that glucocorticoid treatment may affect both endothelial [25, 26] and inducible nitric oxide synthase activity [27, 28]. It is quite plausible that glucocorticoid effects on nitric oxide synthase may be time dependent, requiring transcription and translational mechanisms, making preoperative glucocorticoid therapy more efficacious.
In addition to its role as a pulmonary vasoconstrictor, endothelin-1 has been found to be a proinflammatory cytokine, playing an important role in neutrophil activation. Prior studies have demonstrated that endothelin-1 blockade decreases leukocyte-mediated pulmonary [24] and myocardial injury [29]. Therefore, the ability of glucocorticoid therapy to attenuate endothelin-1 release suggests another potential mechanism by which glucocorticoids might decrease pulmonary ischemia-reperfusion injury and improve pulmonary recovery.
In this study we chose to focus on specific active proteins, such as endothelin-1 and ICAM-1, based on continued evidence of a major regulatory role in the immature piglets. Prior animal and human studies have failed to demonstrate a significant role for TNF- in immature organisms. TNF- was not detectable at the beginning of the experiments in the current study and has not been altered by CPB in prior experiments. We have previously reported clinical and animal data regarding IL-6 and IL-10 [13, 17].
Neutrophil-mediated injury and regulation
Additionally, our data demonstrate attenuation of reperfusion-induced ICAM-1 upregulation with glucocorticoid therapy. Decreased ICAM-1 was associated with lower pulmonary myeloperoxidase activity suggesting less neutrophil adhesion and activation. The decrease in ICAM-1 may be a direct result of decreased NF-B activation secondary to glucocorticoid upregulation of the NF-B inhibitor, IB. Inactivation of NF-B by forming a complex with IB results in decreased NF-B-regulated transcription of pathways in the ischemia and reperfusion injury cascade, including ICAM-1. Although NF-B binding to DNA was not directly measured in this study, other investigators have demonstrated decreased NF-B activity in other organ systems with glucocorticoid therapy [30]. As NF-B is a major regulator of endothelin-1 and ICAM-1 gene activity, it would be expected that stabilization of the NF-B-IB complex would attenuate vasoconstriction and leukocyte-mediated injury. Another proposed mechanism by which glucocorticoids might decrease NF-B activation of pro-inflammatory cytokines is through protein-protein interaction, possibly at gene transcription sites [31].
Decreased ICAM-1 protein was associated with lower myeloperoxidase activity in animals treated with pre and intraoperative steroids and animals receiving intraoperative steroids alone tended toward lower ICAM-1 levels although the decrease did not reach significance (p = 0.064). Similar levels of IB protein in animals treated with intraoperative only and pre and intraoperative glucocorticoids did not translate into the same reduction in ICAM-1 mRNA levels, but did result in similar reductions in myeloperoxidase activity in lung tissue. These results underscore the complexity and recognize the many factors that might affect pulmonary recovery after CPB and DHCA.
In the current study we demonstrated that CPB and DHCA resulted in physiologic pulmonary injury characterized by increased PVR. Furthermore, leukocyte-activation and perturbations in endothlin-1, and likely the nitric oxide pathway, occurred. Glucocorticoid therapy resulted in marked attenuation of endothlein-1 levels that, in the case of preoperative and intraoperative administration, was associated with alleviation of pulmonary hypertension after CPB and DHCA. The absence of improvement in PVR despite lower endothelin-1 in the intraoperative glucocorticoid only group, may possibly be related to additional factors such as NOS activity which was not evaluated in this study. Additionally, glucocorticoid therapy decreased ICAM-1 mRNA and protein, possibly secondary to maintenance of IB levels and decreased NF-B activation. As a result, leukocyte activation, as determined by lower myeloperoxidase activity, was reduced.
|
Acknowledgments |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterial and methodsResultsCommentAcknowledgmentsReferences |
|---|
B DNA binding activity by glucocorticoids in rat brain. Neurosci Lett 1995;198:41-44.[Medline]
B by interfering with serine-2 phosphorylation of the RNA polymerase II carboxy-terminal domain. Genes Dev 2000;14:2314-2329.This article has been cited by other articles:
|
|
 |
|
  D. J. Kozik and J. S. Tweddell Characterizing the Inflammatory Response to Cardiopulmonary Bypass in Children Ann. Thorac. Surg., June 1, 2006; 81(6): S2347 - S2354. [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 |
= no GC; 
= preoperative and intraoperative GC; 
= intraoperative GC. (GC = glucocorticoids.)