Investigation of transcriptional changes underlying calcification of aortic valve

Funding Acknowledgements

Type of funding sources: Foundation. Main funding source(s): The Russian Science Foundation

Introduction

Aortic valve stenosis due to calcification of valve cusps is the most common valve disease in the world today. The main feature of this condition is a progressive mineralization of valve tissue. The mechanisms underlying this process is still unknown, but in recent years it has become clear that pathological mineralization of heart and blood vessels has some similarities with the physiological process of bone formation. It has been suggested that interstitial cells (VICs) are the main functional units in the valve that undergo calcification. However, the early initiating mechanisms that trigger osteogenic transformation of cells remain unclear.

Purpose

The aim of the present study was to elucidate the most responsive time point of osteogenic differentiation induction and to identify the main osteogenic markers that mediate pathological calcification in human aortic valve.

Methods

VICs were obtained from patients with aortic valve calcification and from healthy aortic valves. The effectiveness of cell cultures osteogenic differentiation was estimated by Alizarin Red staining. Investigation of gene expression changes upon osteogenic differentiation was performed by qPCR and RNA sequencing.

Results

We found that 48 hours after the induction of osteogenic differentiation is the most relevant time point to identify the early regulators of osteogenic transformation of the cells. That is the time when the most intensive response from osteogenic markers takes place – BGLAP, OPG, OGN, RUNX2 – in comparison to 24, 72 and 96 hours of differentiation in both patient’s and healthy cells. We found out that induction of osteogenic differentiation on early stages initiates transcriptional program that serve to induce the next molecular events which recruit phenotype-specific osteogenes. We revealed that 558 and 232 genes which were up and down regulated during differentiation were the same for healthy and patient’s cells. However, there was a number of genes which was specific for either patient’s or healthy cells.

Conclusions

We presume that a great amount of the main molecular participants of osteogenic differentiation is shared between different types of cells which are prone to differentiation. However, we perform the results about specificity and difference between the mechanisms of osteogenic differentiation of patient’s and healthy cells.

Intracellular Complement Component 3 Attenuated Ischemia-Reperfusion Injury in the Isolated Buffer-Perfused Mouse Heart and Is Associated With Improved Metabolic Homeostasis

The innate immune system is rapidly activated during myocardial infarction and blockade of extracellular complement system reduces infarct size. Intracellular complement, however, appears to be closely linked to metabolic pathways and its role in ischemia-reperfusion injury is unknown and may be different from complement activation in the circulation. The purpose of the present study was to investigate the role of intracellular complement in isolated, retrogradely buffer-perfused hearts and cardiac cells from adult male wild type mice (WT) and from adult male mice with knockout of complement component 3 (C3KO). Main findings: (i) Intracellular C3 protein was expressed in isolated cardiomyocytes and in whole hearts, (ii) after ischemia-reperfusion injury, C3KO hearts had larger infarct size (32 ± 9% in C3KO vs. 22 ± 7% in WT; p=0.008) and impaired post-ischemic relaxation compared to WT hearts, (iii) C3KO cardiomyocytes had lower basal oxidative respiration compared to WT cardiomyocytes, (iv) blocking mTOR decreased Akt phosphorylation in WT, but not in C3KO cardiomyocytes, (v) after ischemia, WT hearts had higher levels of ATP, but lower levels of both reduced and oxidized nicotinamide adenine dinucleotide (NADH and NAD+, respectively) compared to C3KO hearts. Conclusion: intracellular C3 protected the heart against ischemia-reperfusion injury, possibly due to its role in metabolic pathways important for energy production and cell survival.

P4487Inhibition of aortic valve calcification by SNF472 in vitro

Background

Calcific aortic valve disease is the 2nd most frequent cause of open heart surgery. The valve interstitial cells (VIC) are crucial for calcification. SNF472 (a derivative of phytic acid) is a calcification inhibitor currently in clinical development for the treatment of cardiovascular calcification (Phase 2 CaLIPSO trial, EudraCT 2016–002834–59). SNF472 has been shown to inhibit vascular calcification in several preclinical models.

Purpose

1. Establish a new model of calcification in cultured human VIC; 2. Investigate whether SNF472 would inhibit calcification in this model, and 3. Study if SNF472 might inhibit ongoing calcification processes.

Methods

Healthy and calcified aortic valves were obtained from heart transplant recipients and patients undergoing aortic valve replacement due to calcific valve disease, respectively. VIC were isolated and seeded in basic growth medium, osteogenic differentiation medium (Osteodiff) alone, and with addition of different concentrations of SNF472. The following series of studies were performed: 1. VIC from healthy and calcified valves were cultured for three weeks with Osteodiff; 2. VIC from calcified valves were cultured for 3 weeks in Osteodiff media with 0, 1, 3, 10, 30, or 100 μM SNF472; 3. VIC from calcified valves were cultured for 3 weeks in Osteodiff media in total, but after 1 or 2 weeks 30 or 100 μM SNF472 was added to the cultures (n=8). Calcification was visualized by Alzarin Red staining and quantified by spectrophotometry. Statistics analysis was performed nonparametric One-Way ANOVA (Friedman and Kruskal–Wallis tests) with Dunn's post-test.

Results

Calcification was found to be 30% stronger in cultures of VIC from calcified valves as compared to cultured VIC from healthy valves (p=0.03). SNF472 successfully inhibited VIC calcification in a dose-dependent manner. SNF472 concentrations of 1, and 3 μM inhibited calcification by 7% (not significant) and 66% (p=0.08) respectively. Concentrations of 10, 30, and 100 μM completely inhibited calcification. 30 and 100 μM of SNF472 added after 1 week reduced ongoing calcification by 84% (p<0.01) and 100% (p<0.01) respectively. When given after 2 weeks of ongoing calcification non-significant inhibition was still observed (21 and 30%, respectively).

Conclusions

VIC from calcified valves have a more pro-calcification phenotype than VIC from healthy valves. SNF472 is able to inhibit the development VIC calcification in vitro. By early intervention SNF472 is also able to stop the progression of ongoing calcification. SNF472 shows to be a promising therapy to treat heart valve calcification.

Acknowledgement/Funding

EC FP7 (GA 609020), Balearic Islands Government grant (ES01/TCAI/41_2017), FEDER 2014-2020, Laboratoris Sanifit, Palma, Spain; University of Oslo

SNF472, a novel anti-crystallization agent, inhibits induced calcification in an in vitro model of human aortic valve calcification

The purpose of the present study was to investigate whether SNF472, the hexasodium salt of myo-inositol hexaphosphate (IP6 or phytate): 1. Inhibits induced calcification in cultured aortic valve interstitial cells (VIC) as an in vitro model of aortic valve stenosis and 2. Whether inhibition is different in VIC obtained from healthy and calcified aortic valves. VIC from healthy (n = 5) and calcified (n = 7) human aortic valves were seeded in basic growth medium, osteogenic differentiation medium alone, or in osteogenic medium with SNF472 (3, 10, and 30 μM) and cultivated for 3 weeks. Calcification was quantified spectrophotometrically after Alizarin Red staining. In VIC from calcified valves, a complete inhibition of calcification was observed with SNF472 concentrations of 10 and 30 μM (p < .01), significantly stronger than in VIC from healthy valves. When SNF472 was added to VIC after 1 week in osteogenic medium, 30 and 100 μM SNF472 inhibited the progression of ongoing calcification by 81 and 100% (p < .01), respectively. The same concentrations of SNF472 given after 2 weeks reduced calcification by 35 and 40% respectively (not significant). SNF472 inhibited both the formation and the progression of calcification with the strongest effect in VIC from calcified valves.

Influence of a Concentrated Ethylester Compound of n-3 Fatty Acids on Lipids, Platelets and Coagulation in Patients Undergoing Coronary Bypass Surgery

Twenty patients accepted for coronary bypass surgery were randomized to receive either a concentrated ethylester compound of n-3 fatty acids, with a daily dose of 3.15 g of eicosapentaenoic acid (EPA) and 1.89 g of docosahexaenoic acid (DHA), or corn oil (controls) in a double blind study, to evaluate the effect on lipids, platelets and coagulation during the pre- and postoperative phase.

Only patients with fasting triglyceride (TG) levels ≥1.6 mmol/1 at recruitment were eligible. The study was continued for 5 to 6 months. Surgery was usually performed at mid-intervention. Blood samples were collected during morning hours in fasting subjects, just prior to intervention, preoperatively and at final postoperative follow-up. Moreover, blood loss was accurately accounted for postoperatively.

A threefold increase (p = 0.0001) of EPA was noted at pre-and postoperative follow-up. TG-levels were reduced 20 and 39%, respectively, in patients on n-3 fatty acids, reaching statistical significance at end of intervention (p = 0.034). TG-levels in controls remained largely unchanged. In patients on n-3 fatty acids, there was a statistically significant increase in serum total cholesterol preoperatively, but this change was no longer present at completion of the study.

No significant changes were noted in platelet function, as judged by bleeding time, collagen induced platelet aggregation and release of TxB2 during aggregation. Parameters of extrinsic coagulation, including phospholipase C-sensitive factor VII (PLC-VII) and extrinsic pathway inhibitor (EPI), also remained essentially unchanged in both groups of patients. However, fibrinogen was significantly reduced in controls (p <0.05) at end of intervention. Moreover, a strong positive correlation was noted between PLC-VII and TG (r = 0.77, p = 0.0001).

No significant difference in postoperative bleeding was noted between the two groups of patients.

Lipopolysaccharide Induced Monocyte Thromboplastin Synthesis and Coagulation Responses in Patients Undergoing Coronary Bypass Surgery after Preoperative Supplementation with n-3 Fatty Acids

Twenty patients with coronary heart disease (CHD) and elevated serum lipids were randomized into 2 groups of 10 to receive encapsulated preparations of either a concentrated ethylester form of eicosapentaenoic acid (EPA) and docosa-hexaenoic acid (DHA) or corn oil in doses of 6 g per day, given double blindly for approximately two months prior to coronary bypass surgery. Lipopolysaccharide (LPS) induced monocyte thromboplastin synthesis was studied during the preoperative period and one week following surgery. The ability of n-3 fatty acids to modify tissue factor pathway inhibitor (TFPI) and tissue plasminogen activator inhibitor (PAI-1) was also evaluated along with fibrinogen and thrombin-antithrombin III (TAT) complexes.

No significant changes were noted preoperatively. Monocyte reactivity, PAI-1, fibrinogen and TAT increased significantly after surgery. These changes were not modified by preoperative loading with n-3 fatty acids.

Mode of perfusion influences infarct size, coronary flow and stress kinases in the isolated mouse heart

AIM

The isolated, retrogradely perfused heart (modified Langendorff model) is a widely used method in experimental heart research. The presence of an intraventricular balloon is necessary to get functional measurements. We have previously shown that the balloon induces phosphorylation of some suggested cardioprotective mitogen-activated protein kinases (MAPK): P38-MAPK, ERK 1/2 and JNK. We hypothesized that the balloon could influence cardioprotection, protect against ischaemia reperfusion injury and interfere with coronary flow.

METHODS AND RESULTS

Isolated mouse hearts were perfused for 5, 10, 20, 40 and 60 min with a balloon in the left ventricle. We found a wavelike phosphorylation of all MAPK while AKT displayed a gradual dephosphorylation when compared to non-perfused hearts. Hearts were subjected to 20 min of stabilization with or without the balloon, followed by 35 min of ischaemia and 120 min of reperfusion. Although the MAPK were phosphorylated, the infarcts were larger in the balloon group. When the balloon was present during the entire protocol, compared to removal at the end of ischaemia, the infarct size was also larger, especially in the endocardial layer. The balloon reduced post-ischaemic endocardial coronary flow, despite a higher average flow, indicating a hyperperfused epicard. Blocking the balloon-induced ERK 1/2 phosphorylation during stabilization did not affect infarct size. The effect of post-conditioning was influenced by the balloon, showing reduced infarct size when the balloon was present.

CONCLUSION

The balloon used for pressure measurements may contributes to cell death possibly by reducing endocardial coronary flow.

Leucocytes and cardiopulmonary bypass: in vitro production of oxygen free radicals and trapping in the reperfused myocardium

The production of oxygen free radicals (OFR) by leucocytes was evaluated ex vivo by chemiluminescence (CL) before, during and after routine coronary artery bypass surgery (group A, n=11). The possibility of leucocyte trapping in the coronary circulation during the early reperfusion period was also investigated (group B, n=9). In group A, arterial blood samples were taken immediately before the start of surgery during anaesthesia, five minutes before and five and 30 minutes after the start of cardiopulmonary bypass (CPB), five minutes before and five and 30 minutes after the start of reperfusion of the heart, and then four and 24 hours after the end of CPB. In group B, arterial and coronary sinus blood samples were simultaneously drawn five and 30 minutes after the release of the aortic crossclamp. All blood samples were corrected for haemodilution. In group A, both CL and the level of circulating leucocytes declined during CPB. The lowest value of CL was measured 30 minutes after the start of CPB (69± 2% of baseline values) (mean±SEM). The lowest level of leucocytes was found after 30 minutes of CPB: 2.6±0.4 (109/l) vs 4.2±0.5 before surgery. Twenty-four hours after CPB, CL was increased to 170±49% and a leucocytosis was present (12.2±1.1). In group B, after five minutes of reperfusion the number of circulating leucocytes in arterial blood was 3.8±0.9 x 10 9/l as compared to 2.2±0.5 x 109/l in the coronary sinus (p<0.0017). However, no such difference was found after 30 minutes of reperfusion. The decreased CL during CPB was probably due to in vivo activation and exhaustion of leucocytes. The postischaemic trapping of these cells may play a pathogenetic role in reperfusion injury.

Mitochondrial DNA damage and repair during ischemia-reperfusion injury of the heart

Ischemia-reperfusion (IR) injury of the heart generates reactive oxygen species that oxidize macromolecules including mitochondrial DNA (mtDNA). The 8-oxoguanine DNA glycosylase (OGG1) works synergistically with MutY DNA glycosylase (MYH) to maintain mtDNA integrity. Our objective was to study the functional outcome of lacking the repair enzymes OGG1 and MYH after myocardial IR and we hypothesized that OGG1 and MYH are important enzymes to preserve mtDNA and heart function after IR. Ex vivo global ischemia for 30min followed by 10min of reperfusion induced mtDNA damage that was removed within 60min of reperfusion in wild-type mice. After 60min of reperfusion the ogg1(-/-) mice demonstrated increased mtDNA copy number and decreased mtDNA damage removal suggesting that OGG1 is responsible for removal of IR-induced mtDNA damage and copy number regulation. mtDNA damage was not detected in the ogg1(-/-)/myh(-/-), inferring that adenine opposite 8-oxoguanine is an abundant mtDNA lesion upon IR. The level and integrity of mtDNA were restored in all genotypes after 35min of regional ischemia and six week reperfusion with no change in cardiac function. No consistent upregulation of other mitochondrial base excision repair enzymes in any of our knockout models was found. Thus repair of mtDNA oxidative base lesions may not be important for maintenance of cardiac function during IR injury in vivo. This article is part of a Special Issue entitled "Mitochondria: From Basic Mitochondrial Biology to Cardiovascular Disease."

Proteasome inhibitors eliminate protective effect of postconditioning in cultured neonatal cardiomyocytes

A role of proteasomal proteolysis in the pathogenesis of ischemia-reperfusion is being actively studied. To evaluate the participation of the proteasome in postconditioning phenomenon, we used primary culture of neonatal cardiomyocytes. 30 minutes of anoxia followed by 60 minutes of reoxygenation was undergone. Postconditioning was modeled by 3 cycles of 1-minute reoxygenation followed by 1-minute anoxia, respectively. Clasto-lactacystin b-lactone, a specific proteasome inhibitor, in the dose that does not cause cell death (2.5 mM) was added to the culture medium just before the cycles of postconditioning. Percentages of living, necrotic, and apoptotic cells were determined by staining with bisBenzimide and propidium iodide. Autophagy was demonstrated by staining vacuolar structures with monodansyl cadaverine. Proteasomal activity was determined by cleavage intensity of specific fluorogenic substrates. Trypsin-like, chymotrypsin-like and peptidyl-glutamyl peptide-hydrolyzing (PGPH) activities were decreased after anoxia. Reoxygenation led to an increase in trypsin-like and chymotrypsin-like activities comparing to anoxia, but these parameters never reached the control levels. PGPH activity was restored up to the initial level. Postconditioning increased numbers of living cells and decreased that of necrotic, apoptotic and autophagic cells. Paradoxically, it was established, that proteasome inhibitors prevented the necrotic and apoptotic cell death of cardiomyocytes in anoxia-reoxygenation, but in the same concentration abolished the effects of postconditioning. The data obtained permit to suppose that proteasome inhibitors can be used for pharmacological postconditioning.

Postconditioning prevents apoptotic necrotic and autophagic cardiomyocyte cell death in culture

In the paper the data concerning the possibility of the reproduction of the postconditioning phenomena in the cardyomicyte culture are presented. Primary cultures of cardiomyocytes from neonatal rats underwent 30 minutes of anoxia followed by 60 minutes of reoxygenation. Three different models of postconditioning were used: 3 cycles of 1, 3, or 5 minutes of reoxygenation followed by 1, 3, or 5 minutes of anoxia, respectively. The percentage of living, necrotic, and apoptotic cells were determined by staining with Hoechst 33342 and propidium iodide. Autophagy was demonstrated by the staining of vacuolar structures in vivo by monodansyl cadaverine. After anoxia and reoxygenation the amount of living, necrotic and apoptotic cells were 79 +/- 1.5, 7.8 +/- 0.9 and 13 +/- 1.5 %, respectively (in unstimulated cell culture 90 +/- 0.8, 3.3 +/- 0.3, and 5.5 +/- 0.7, P < 0.0001 for all). Postconditioning with 1 min anoxia 3-fold increased the amount of living cells and decreased the number of necrotic and apoptotic cells (P = 0.002, P = 0.02 and P = 0.043 respectively). Postconditioning with cycles of 3 and 5 minutes had a gradually reduced effect compared to cycles of 1 minute. The percentage of autophagic cells in control cell culture was 4.3 +/- 0.3%. This number increased after anoxia-reoxygenation to 14 +/- 0.8%, and was reduced by postconditioning (P < 0.001). The data obtained indicate that postconditioning is one of the effective methods of cardioprotection and could effectively decrease the amount of cardiomyocytes with traits of programmed or non-programmed cell death.

Importance of preanalytical handling of samples for measurement of cardiac troponin T in coronary effluent from isolated rat hearts

The isolated, buffer-perfused heart is probably the most widely used model in experimental heart research, and the coronary effluent is often analysed for markers of myocardial injury. Adsorption to surrounding materials may be a serious problem of protein measurements in solutions with low protein concentrations. The aims of the present study were to investigate the importance of the preanalytical phase when measuring cardiac troponin T (cTnT) in a buffer perfusate and to investigate whether addition of albumin to the effluent might increase recovery of cTnT and improve the assay. Coronary effluent was collected in tubes of different materials and in tubes with 40 g/L bovine albumin, and then frozen. cTnT was analysed at different time points after withdrawal from the freezer. cTnT was 2.3-119 times higher in effluent with albumin. In effluent without albumin, cTnT concentration declined to 2% of the initial concentration after two episodes of freezing and thawing. The cTnT loss could not be prevented by using polystyrene or siliconized glass, but was partially inhibited in effluent with albumin. Furthermore, creatine kinase and lactate dehydrogenase levels were higher in effluent with albumin. The within-series coefficient of variation for cTnT was markedly improved when using effluent with albumin.

Effects of sex, gonadectomy, and oestrogen substitution on ischaemic preconditioning and ischaemia-reperfusion injury in mice

AIM

Ischaemic preconditioning (IPC) has been demonstrated to protect heart function and viability, but has been predominantly studied in male animals.

METHODS

We studied a possible influence of sex and oestrogen for protection in IPC. Infarct size and heart function after 40 min global ischaemia and 60 min reperfusion with or without preceding classic IPC was investigated in Langendorff-perfused hearts. Hearts were harvested from 10-week-old male and female C57BL6 mice with or without gonadectomy 6 weeks earlier, or gonadectomy and substitution with 17 beta-oestradiol for 4 weeks (n = 104).

RESULTS

Classic IPC reduced depression of left ventricular developed pressure (P < 0.01), attenuated the increase of end-diastolic pressure (P < 0.01), and reduced infarct size (P < 0.01) in hearts of untreated male mice, but failed to protect untreated females which had improved functional recovery and smaller infarctions than untreated males. After gonadectomy of female mice, developed pressure was reduced (P < 0.01) and infarct size increased (P < 0.01) compared with normal females, with no protection of preconditioning. The changes were not reversed by 17 beta-oestradiol substitution. In hearts of gonadectomized males, the post-ischaemic increase of end-diastolic pressure was attenuated (P < 0.01), and enhanced after substitution with 17 beta-oestradiol (P < 0.01). The preconditioning effect disappeared after gonadectomy and gonadectomy with substitution in male mice.

CONCLUSION

There is a sex difference in evoking preconditioning in male and female mice which is only partially dependent on sex hormones.

Exposure of rats to hyperoxia enhances relaxation of isolated aortic rings and reduces infarct size of isolated hearts

Exposure of rats to hyperoxia before organ harvesting protected their isolated hearts against global ischaemia-reperfusion injury in a previous study. The present study investigates whether hyperoxia influences vasomotor function and regional ischaemia of the heart. Isolated rings of the thoracic aorta were obtained from rats immediately or 24 h after in vivo exposure to 60 min of hyperoxia (>95% O2), and the in vitro dose-response to phenylephrine (PHE), prostaglandin F2alpha (PGF2alpha) and endothelin-1 (ET-1), acetylcholine (Ach) and sodium nitroprusside (SNP) was assessed. Hyperoxia in vivo increased the relaxation of aortic rings to Ach and SNP, while it delayed contraction to PHE. The effect was more evident when the vessels were harvested immediately rather than 24 h after hyperoxic exposure. In separate experiments rat hearts were isolated immediately after hyperoxia, buffer-perfused, and subjected to 30 min of regional ischaemia and reperfused for 120 min. Infarct size was determined by triphenyl tetrazolium chloride staining. Hyperoxia significantly reduced infarct size. In normoxic controls 23.0 +/- 8.3% of the area at risk was infarcted, while in hyperoxic animals infarct size was 14.8 +/- 5.6% of the area at risk (P = 0.012). Exposure of rats to hyperoxia modifies the vasomotor response of isolated aortic rings, and reduces the infarct size of isolated rat heart. These novel aspects of hyperoxic treatment require further studies to explore the potential of its clinical application.

The effect of a low molecular weight inhibitor of lipid peroxidation on ultrastructural alterations to ischemia-reperfusion in the isolated rat heart

The effects of H290/51, a novel indenoindole derivative inhibitor of lipid peroxidation, on ultrastructural changes during cardiac ischemia-reperfusion injury were investigated. Langendorff-perfused rat hearts were exposed to 30 minutes of global ischemia followed by 20 minutes of reperfusion: Group A: Control hearts with standard buffer perfusion with vehicle added. Group B: H290/51 (10(-6) mol/l) added to buffer throughout stabilisation and reperfusion. In an additional Group C, where hearts were given H290/51, but not subjected to ischemia, the ultrastructure was preserved till the end of reperfusion. Absolute volumes and calculated volume fractions (Vv) of tissue and subcellular components were assessed with quantitative stereologic morphometry. After ischemia the increase in volume of extracellular interstitium was inhibited by H290/51 (247 +/- 80 vs. 159 +/- 50 microl, mean +/- SD, groups A and B, respectively, p<0.05). The Vv (interstitium/myocard) was higher in control hearts (0.318 +/- 0.062 vs. 0.206 +/- 0.067, p<0.05). Vv (cell edema/myocyte) was higher in the control group (0.144 +/- 0.07 vs. 0.083 +/- 0.033, p<0.05). Vv (myocyte/myocard) was higher in group B after ischemia than in the control group (0.622 +/- 0.071 vs. 0.707 +/- 0.052, p<0.05). The decreased Vv (capillary/myocard) after ischemia was inhibited by H290/51. After reperfusion there was no difference between groups. Treatment with H290/51 reduced edema and ensured better preserved sarcolemmal membrane structure during ischemia. The effect was no longer present after reperfusion.

Cardioprotection by breathing hyperoxic gas-relation to oxygen concentration and exposure time in rats and mice

OBJECTIVES

Breathing a hyperoxic gas (> or =95% O(2)) protects against ischaemia-reperfusion injury in rat and mouse hearts. The present study investigated how oxygen concentration and duration of hyperoxic exposure influenced cardioprotection, and whether hyperoxia might induce delayed cardioprotection (after 24 h).

METHODS

Animals were kept in normal air or in a hyperoxic environment, and their hearts were isolated and Langendorff-perfused immediately or 24 h thereafter. Global ischaemia was induced for 25 min in rats and 40 min in mice, followed by 60 min of reperfusion. Infarct size was determined by triphenyl tetrazolium chloride staining.

RESULTS

In rats exposure to > or =95, 80, and 60%, but not to 40% of oxygen immediately before heart isolation and perfusion improved postischaemic functional recovery. Eighty or more percent of oxygen also reduced infarct size. A preconditioning-like effect could be evoked by 60 or 180 min of hyperoxia, giving both immediate and delayed protection. In the mouse heart protection could be induced by pretreatment for 15 or 30, but not by 60 min with > or =95% oxygen. The protective effect of hyperoxia in mice could be evoked in the immediate model only.

CONCLUSIONS

Hyperoxia protects the isolated rat and mouse heart against ischaemia-reperfusion injury, but some species-different responses exist. The protection depends on both oxygen concentration in inspired air, and duration of hyperoxic exposure.

Postoperative mediastinitis in cardiac surgery - microbiology and pathogenesis

OBJECTIVE

During 1992-2000, postoperative mediastinitis developed after 126 (1.32%) of 9557 consecutive cardiac surgery procedures. The study was done to describe the variation in clinical characteristics and microbiological etiology in mediastinitis.

METHODS

The records of 126 cases of postoperative mediastinitis were reviewed.

RESULTS

The median time from operation to the development of mediastinitis was 7 days. Sternal dehiscence was seen in 86 patients (68%). Coagulase negative staphylococci (CNS) were isolated in 46% of the cases with a verified microbiological etiology, Staphylococcus aureus in 26% and gram-negative bacteria in 18%. CNS were more frequently isolated in patients with sternal dehiscence (44/80, 55%) than in patients with stable sternum (10/38, 26%) (P=0.003). However, S. aureus was more frequent in patients with stable sternum (18/38, 47%) than in patients with sternal dehiscence (13/80, 16%) (P<0.001). High body mass index was associated with coagulase negative staphylococci (P<0.001) and with sternal dehiscence (P=0.008). Chronic obstructive pulmonary disease was also associated with sternal dehiscence (P<0.001) and with coagulase negative staphylococci (P=0.04). Patients who had been reoperated before onset of mediastinitis tended to have an increased risk for a gram-negative etiology (32 vs. 15% in patients not reoperated, P=0.06). The overall 90-day all cause mortality in patients with mediastinitis was 19%. High age, need for reoperation before mediastinitis, and a long primary operation time was associated with increased mortality (P=0.02, P=0.007 and P=0.001, respectively). No specific bacterial etiology was associated with increased mortality nor was the presence of bacteriemia.

CONCLUSIONS

Three different types of postoperative mediastinitis can be distinguished: (1) mediastinitis associated with obesity, chronic obstructive pulmonary disease, and sternal dehiscence, typically caused by coagulase negative staphylococci; (2) mediastinitis following peroperative contamination of the mediastinal space, often caused by S. aureus, and (3) mediastinitis mainly caused by spread from concomitant infections in other sites during the postoperative period, often caused by gram negative rods. The proposed classification of mediastinitis into three groups with different pathogenic mechanisms may be useful in understanding which prophylactic counter measures have the potentials to be effective in a given situation.

Better preserved pulmonary endothelium-dependent vasodilation with off-pump coronary surgery

OBJECTIVE

To investigate if endothelium-dependent vasodilation in the pulmonary circulation was better maintained after off-pump coronary artery bypass grafting (CABG). An impaired pulmonary vascular response to acetylcholine has been observed after cardiopulmonary bypass (CPB) in children, adults and experimentally.

DESIGN

Fourteen patients operated off-pump were compared with 21 patients undergoing conventional CABG with CPB. The indexed pulmonary vascular resistance was measured before and during an infusion of acetylcholine, aiming at a concentration of 10(-6) mol/l in the pulmonary artery. Twelve patients operated on-pump received saline instead of acetylcholine.

RESULTS

Before surgery pulmonary vascular resistance decreased during infusion of acetylcholine by 28% and 25% in the off-pump and on-pump groups. After surgery the decrease was 16% and 6%, respectively (p = 0.028 and p < 0.001, compared to preoperative response). The response did not differ between the two groups before, but did so after surgery (p = 0.01). Saline had no effect.

CONCLUSION

The better maintained endothelium-dependent vasodilation in the off-pump group indicated less endothelial dysfunction.

Fewer reoperations and shorter stay in the cardiac surgical ward when stabilising the sternum with the Ley prosthesis in post-operative mediastinitis

OBJECTIVE

Using the Ley prosthesis, a 0.5mm thick titanium alloy plate for stabilising the sternum, is a new method in the treatment of mediastinitis after open-heart surgery. We report a retrospective analysis of our experience with this device.

METHODS

One hundred consecutive cases of post-operative mediastinitis in the period 1992-1997 were reviewed. The primary procedure at reoperation for infection was as follows: 52 patients were treated with the Ley prosthesis and 48 patients underwent other conventional procedures. The choice of the surgical technique depended on the attending surgeon. The prosthesis was used more frequently in patients with sternal dehiscence (P<0.001) but otherwise patients' characteristics were similar in the two groups. As a control population for outcome data, 100 uninfected patients were matched with regard to operative procedure, age, sex and date of surgery.

RESULTS

The median hospital stay was 48.5 days in the mediastinitis group vs. 14 days in the control group. The all-cause 90-day mortality in the mediastinitis group was 18% vs. 5% in the control group. The 52 patients treated with the Ley prosthesis had a median length of stay in the cardiac surgery ward for 29 days vs. 41.5 days in the mediastinitis group not treated with the prosthesis (P=0.013). However, when the total length of stay including hospitals outside the cardiac surgery ward was taken into account, the prosthesis did not reduce the length of stay. Only 8/52 patients treated with the prosthesis required further surgery vs. 23/48 patients who were not primarily treated with the prosthesis (P<0.001). The Ley prosthesis had no impact on mortality.

CONCLUSION

The Ley prosthesis is a valuable adjunct to the treatment of mediastinitis after open-heart surgery. A shorter stay at the cardiac surgery ward and a reduced need for further surgical procedures were observed when using this prosthesis.

Increase in serum S100A1-B and S100BB during cardiac surgery arises from extracerebral sources

BACKGROUND

Elevated levels of serum S100B after coronary artery bypass grafting may arise from extracerebral contamination. Serum S100B content was analyzed in several tissues, and the two dimers S100A1-B and S100BB were analyzed separately in blood.

METHODS

Serum, shed blood, marrow, fat, and muscle were studied in patients undergoing coronary artery bypass grafting with cardiopulmonary bypass using suction either to the cardiotomy reservoir (group 1, n = 10) or to a cell-saving device (group 2, n = 10), or operated on off-pump (group 3, n = 10).

RESULTS

Serum S100B was sixfold higher in group 1 than in groups 2 and 3, which were identical. The same ratio between S100A1-B and S100BB was found in all groups. When compared with serum, S100B was 10(2) to 10(4) times higher in marrow, fat, muscle tissue, and shed blood.

CONCLUSIONS

Separate analysis of S100A1-B and S100BB did not distinguish between S100B of cerebral and extracerebral origin. The concept that S100B only originates in astroglial and Schwann cells is wrong. Fat, muscle, and marrow in mediastinal blood contain high levels of S100B. Cardiopulmonary bypass caused no increase in S100B.

Preconditioning protects the severely atherosclerotic mouse heart

BACKGROUND

Coronary atherosclerosis has profound effects on vascular and myocardial biology, and it has been speculated that the atherosclerotic heart does not benefit from ischemic preconditioning.

METHODS

To investigate if atherosclerosis would influence the preconditioning response, Apolipoprotein E/low density lipoprotein (LDL) receptor double knockout mice (ApoE/LDLr-/-) were fed an atherogenic diet (21% fat, 0.15% cholesterol) for 6 to 8 months. At that time, extensive atherosclerotic lesions throughout the coronary tree were seen in transverse sections stained with Oil Red-O. Hearts of ApoE/LDLr-/- mice were Langendorff-perfused with 40 minutes of global ischemia and 60 minutes reperfusion, and compared with C57BL/6 controls. Preconditioning with two episodes of 2 minutes of ischemia and 5 minutes reperfusion, or exposing the mice to a hyperoxic environment (O2 > 98%) for 60 minutes before heart perfusion, was performed.

RESULTS

Hearts of mice with coronary atherosclerosis had worse postischemic function, and increased infarct size and troponin T release compared to hearts of C57BL/6 mice. Ischemic preconditioning improved postischemic ventricular function, and reduced myocardial infarct size and troponin T release in both normal and ApoE/LDLr-/- mice. The effects were most pronounced in ApoE/LDLr-/- hearts. Exposure to hyperoxia exerted a similar protection of function and cell viability of ApoE/LDLr-/- mice hearts.

CONCLUSIONS

These findings suggest that the severely atherosclerotic heart may be protected by preconditioning induced by ischemia or hyperoxia.

Correlation between a mid-ventricular volume segment and global left ventricular volume measured by the conductance catheter

OBJECTIVES

To investigate whether acute volume changes in single volume segments of the left ventricle can be correlated with global volume changes. If so, changes in global volume might be predicted from changes in segmental volumes.

DESIGN

Volume changes were recorded in six pigs in five intraventricular segments, from apex to heart base, using the conductance catheter (at baseline, after 60 min of apical ischaemia, during preload reduction and afterload increase). A computer algorithm was created to calculate the instantaneous absolute difference between the curve shape of global and normalized segmental volume as a percentage of global stroke volume.

RESULTS

For a mid-cardiac volume segment constituting 34 (14-39)% [median (range)] of global stroke volume, the mean difference over a cardiac cycle was 4 (1-8)% at baseline. Apical ischaemia resulted in apical dyskinesia, but did not influence the mid-cardiac segment.

CONCLUSIONS

The volume curve from a segment at mid-cardiac level seems to be a good estimator of the global volume curve, thus giving a foundation for estimation of global volume changes from such a segment.

Pretreating rats with hyperoxia attenuates ischemia-reperfusion injury of the heart

Oxidative stress may precondition the heart. The present study investigated whether hyperoxia elicits a preconditioning-like response. Rats were kept in a hyperoxic (>95% O2) environment for 60 or 180 minutes. Hearts were Langendorff-perfused immediately or 24 hours after hyperoxia, and exposed to 25 minutes of global ischemia and 60 minutes of reperfusion. Whole blood was sampled after 60 and 180 minutes of hyperoxia for oxidative stress markers. Hearts were sampled immediately or 24 hours after hyperoxia for measurement of antioxidants, lipid peroxidation products, heat shock protein 72 and endothelial nitric oxide synthase. At the end of reperfusion after 1 h hyperoxia, infarct size was determined by tetrazolium staining. Hyperoxia increased serum levels of conjugated dienes, reduced serum antioxidative protection, reduced reperfusion arrhythmias in most groups, and improved myocardial function. Infarct size was reduced from 45% of myocardial tissue in controls to 22% in treated animals. The myocardial activity of antioxidant enzymes, content of heat shock protein 72, and endothelial nitric oxide synthase in myocardial tissue were not influenced. In conclusion, hyperoxia induces a low-graded systemic oxidative stress, improves postischemic cardiac function and reduces infarct size. The mediators of protection remain to be determined.

Induction of inflammatory mediators during reperfusion of the human heart

BACKGROUND

Cardioplegia and reperfusion may induce an inflammatory reaction, which may contribute to postoperative morbidity and mortality.

METHODS

Gene expression of cytokines, adhesion molecules, and vasoactive substances was evaluated in left ventricular biopsies taken before cardioplegia (lasting approximately 70 minutes) and after reperfusion (approximately 40 minutes) from 19 patients (5 with valvular or combined disease, 7 with stable angina pectoris, 7 with unstable angina). mRNA was extracted and amplified with a semiquantitative reverse transcription polymerase chain reaction.

RESULTS

Cardioplegia-reperfusion increased mRNA for E-selectin by a factor of 17 +/- 5 (p < 0.002) (mean +/- SEM), interleukin-1beta, with 9 +/- 3 (p < 0.007), tumor necrosis factor-alpha with 6 +/- 3 (p < 0.05), interleukin-2 receptor alpha chain CD25 with 2 +/- 0.6 (p < 0.04), and intercellular adhesion molecule-1 with 2 +/- 0.4 (p < 0.005). Before cardioplegia, mRNA for endothelial nitric oxide synthase was predominantly detected in unstable angina patients, and increased by a factor of 11 +/- 6 (p < 0.02) during reperfusion. mRNA for endothelin-1 decreased by a factor of 0.5 +/- 0.1 (p < 0.0005). The changes were more pronounced in unstable patients. The transcription factor nuclear factor kappa B (NFkappaB), which regulates expression of inflammatory mediators, was activated during reperfusion (n = 10, p < 0.0001).

CONCLUSIONS

Open heart surgery induces an inflammatory response in the human heart, which is more pronounced in patients with unstable angina. It involves NFkappaB activation and expression of several NFkappaB-regulated genes.

Neurologic injury in cardiac surgery: aortic atherosclerosis emerges as the single most important risk factor

With older and sicker patients undergoing cardiac surgery, neurologic injury has emerged as an increasingly important cause of rising costs, morbidity and mortality. Several studies investigating the relationship between atherosclerotic aortic disease and subsequent adverse clinical outcomes have demonstrated that the single most important risk factor for neurologic injury following cardiac surgery is the presence of aortic atheromatous disease. The results of these studies suggest that atheroemboli are correlated with neurologic injury following cardiac surgery. Surgical techniques to avoid and prevent particulate debris during cardiac surgery may be a major step in preventing severe neurologic injury.

Glucocorticoid pretreatment protects cardiac function and induces cardiac heat shock protein 72

Acute administration of glucocortiocoids reduces inflammation. Increasing knowledge of the mechanisms of action indicate that pretreatment with glucocorticoids could have organ-protective effects. We investigated whether pretreatment with methylprednisolone (MP) protected the heart against ischemia-reperfusion dysfunction, and we hypothetized that this protection might be due to induction of the cardioprotective heat shock protein 72 (HSP72). Rats were given vehicle or MP-40 mg/kg im as a double injection starting either 24 or 120 h (5 days) before their hearts were excised for Langendorff perfusion (n = 6-11 hearts in each group). MP improved left ventricular function and coronary flow during reperfusion after 30 min of global ischemia and reduced infarct size. Cardiac HSP72 gradually increased in a 24-h time course after MP treatment, and the increase was sustained 5 days afterward (immunoblotting). HSP72 mRNA was either reduced or unchanged, indicating a posttranscriptional regulation. Pretreatment with hydrocortisone or dexamethasone (n = 7-8 hearts of each) similarily increased cardiac HSP72 24 h afterward. This paper demonstrates that glucocorticoids increase cardiac HSP72 and protect organ function against ischemia-reperfusion injury.

Gene expression of inflammatory mediators in different chambers of the human heart

BACKGROUND

Inflammatory genes may be unevenly expressed in different heart chambers.

METHODS

Biopsies were taken simultaneously from the right atrium (RA), left atrium (LA), and left ventricle (LV) of 19 patients before cardioplegic arrest during open heart surgery. The mRNA expression of tumor necrosis factor alpha (TNFalpha), interleukin 1beta (IL-1beta), inducible and endothelial nitric oxide synthase (iNOS and eNOS), endothelin-1 (ET-1), E-selectin (CD62E), intercellular adhesion molecule-1 (ICAM-1) and its ligand CD18, and CD25 was evaluated with semiquantitative reverse transcription-polymerase chain reaction (RT-PCR).

RESULTS

Expression of TNFalpha mRNA was higher in RA than LA and LV (p<0.05), whereas IL-1beta was more expressed in LA than RA (p<0.05), which was higher than LV (p<0.0001). There were no significant regional differences in the expression of ICAM-1, CD62E, CD25, iNOS, and eNOS. CD18 was higher in RA than LA (p<0.05); ET-1 was more expressed in RA than LV (p<0.04). Patients with stable angina had no expression of eNOS.

CONCLUSIONS

Gene expression of inflammatory mediators was detected in the hearts of patients with different cardiovascular disorders, and was unevenly distributed in different heart chambers. Cardiac biopsies should be taken from the same site.

Unstable angina activates myocardial heat shock protein 72, endothelial nitric oxide synthase, and transcription factors NFkappaB and AP-1

OBJECTIVE

Unstable angina may improve the clinical outcome of acute myocardial infarction, but increases the morbidity and mortality of open heart surgery. We hypothetized that unstable angina influences the myocardium, and investigated the expression of the inducible heat shock protein 72 (HSP72), constitutive HSP73, and endothelial nitric oxide synthase (eNOS), and activation of the transcription factors NFkappaB and AP-1 in cardiac tissue.

METHODS

Biopsies were taken from the right atrium of 15 patients with unstable and 15 with stable angina undergoing coronary artery bypass grafting. Immunoblotting with monoclonal antibodies against HSP72, HSP73, and eNOS were performed on protein extracts, while nuclear proteins were assessed by electromobility shift assay.

RESULTS

When calculating the optical density of the bands, patients with unstable angina had more than twice as much HSP72 and eNOS as stable patients (P<0.005), while HSP73 was similar in both groups. Nuclear translocation of NFkappaB and AP-1 was found in patients with anginal pain shortly before surgery, but not in stable patients or in patients without symptoms for 4 days or more prior to surgery.

CONCLUSIONS

HSP72 and eNOS, which may be associated with cardioprotection in ischemic preconditioning, are increased in atrial tissue of patients with unstable angina. Activation of NFkappaB and AP-1, which regulate a battery of inflammatory genes, was found in hearts of unstable patients. NFkappaB activation may induce a myocardial proinflammatory state, possibly making the unstable myocardium more susceptible to the inflammation induced by open heart surgery.

Pretreatment with methylprednisolone protects the isolated rat heart against ischaemic and oxidative damage

Methylprednisolone (MP), a synthetic glucocorticoid, is widely used clinically and experimentally as acute antiinflammatory treatment. The molecular actions of MP indicate that pretreatment with this drug may be cardioprotective. We investigated if giving rats MP prior to excising their hearts for Langendorff-perfusion protected cardiac function against oxidative stress, and if this was mediated by increasing antioxidant defence or influencing myocardial nitric oxide synthase (NOS). Rats (n=6-11 in each group) were injected with MP (40mg/kg i.m.) or vehicle 24 and 12 h before Langendorff-perfusion with 30 min global ischaemia and 60 min reperfusion, or 10 min perfusion with 180 micromol/L hydrogen peroxide. Other hearts were exposed to 30 min global ischaemia 5 days after MP-injection. Additional hearts were sampled before, during, and after ischaemia for analyzing tissue activity of antioxidant enzymes. Tissue endothelial and inducible NOS (eNOS and iNOS) were investigated by immunoblotting and semiquantitative RT-PCR in a time-course after MP injection. Pretreatment with MP improved left ventricular function and increased coronary flow during postischaemic reperfusion, and this effect was sustained 5 days afterwards. When exposing hearts to hydrogen peroxide, MP improved coronary flow. Catalase, glutathione peroxidase, and oxidized glutathione were increased during reperfusion of MP-treated hearts compared to vehicle only. MP did not influence eNOS at protein or mRNA level. iNOS could not be detected by immunoblotting, indicating low cardiac enzyme content. Its mRNA initially increased the first hour after injection, thereafter decreased. In conclusions, pretreating rats with MP protects the heart against ischaemia-reperfusion dysfunction. This effect could be due to increase of tissue antioxidant activity during reperfusion. MP did not influence cardiac eNOS. mRNA for iNOS was influenced by MP, but the corresponding protein could not be detected.

Effects of cardiac surgery on some clinically used inflammation markers and procalcitonin

One hundred and ten patients were investigated prospectively in a study aimed at creating reference curves for inflammation markers (serum C-reactive protein (CRP), blood leukocyte count, iron, transthyretin and procalcitonin). Blood samples were taken daily and the patients were monitored for signs of infection. Ninety-six patients had no postoperative infections. CRP and leukocyte counts peaked on the third and second postoperative days, respectively. Neither patients operated on off-pump (n = 4) nor patients with minor infections (n = 11) differed from the non-infected group. Two out of three patients with major postoperative infection exhibited a secondary peak in CRP and leukocyte count. Iron and transthyretin decreased initially, followed by a slow increase without any difference between the groups. Procalcitonin was high in some non-infected patients and low in some infected patients. CRP and leukocyte count had a predictable course with a secondary peak in major infections but the other markers did not provide any valuable information.

The effect of cardiotomy suction on the brain injury marker S100beta after cardiopulmonary bypass

BACKGROUND

An increase of S100beta in serum during cardiopulmonary bypass (CPB) has been interpreted as a sign of brain injury. Cardiotomy suction may cause fat embolization, and its role in the S100beta increase was examined.

METHODS

Twenty coronary artery operation patients were randomly assigned to two groups, 10 with suction during CPB to cardiotomy reservoir (CR), 10 to cell saving device (CS). S100beta was measured (immunoassay) in blood from the patients and from cell saving device after processing. In 7 additional patients S100beta was measured in the cell saving device before processing and directly from the wound at sternotomy.

RESULTS

Before anesthesia, serum S100beta was 0.03+/-0.06 microg/L. At the end of CPB it was 2.47+/-1.31 microg/L and 0.44+/-0.27 microg/L (CR vs CS; p < 0.001). S100beta was 33+/-12 microg/L in CS reservoir and 42+/-18 microg/L in blood from the wound.

CONCLUSIONS

Most serum S100beta after CPB with cardiotomy suction may be of extracerebral origin. S100beta after CPB with cell saving device was the same as after off-pump operation. The interpretation that an increase in S100beta during CPB in patients reflects cerebral injury must be questioned.

Increased extracellular brain water after coronary artery bypass grafting is avoided by off-pump surgery

OBJECTIVE

To determine if coronary artery bypass graft (CABG) surgery without cardiopulmonary bypass (CPB) avoids the brain swelling known to occur after CPB, to quantify these brain water compartment changes, and to identify the water shifts as due to intracellular or extracellular water.

DESIGN

Prospective, controlled, and blinded.

SETTING

Cardiac surgical unit in a university teaching hospital.

SUBJECTS

Patients scheduled for CABG who were assigned to conventional (n = 10) or off-pump (n = 7) surgery according to their coronary anatomy.

INTERVENTIONS

Magnetic resonance imaging (MRI) examinations were performed 1 day before surgery and 1 hour and 1 week after CABG surgery.

MAIN OUTCOME MEASURES

Extracellular and intracellular water homeostasis was described quantitatively by calculating the averaged apparent diffusion coefficient of brain water using diffusion-weighted MRI. Blinded visual ordering of the images from the three examinations was performed according to brain size using conventional MRI.

RESULTS

The average diffusion coefficient of brain water increased 4.7%+/-1.5% immediately after CABG with CPB and normalized after 1 week but did not change after CABG without CPB. No focal ischemic changes were seen in either group, and no gross neurologic deficits were observed. Visual analysis showed consistent brain swelling after CPB and variable changes in those operated without CPB.

CONCLUSION

Changes consistent with increased extracellular brain water seen after CABG with CPB were not observed in patients undergoing CABG without CPB. The clinical significance of brain water changes and increased brain water content after surgery with CPB remains undefined.

Post-ischaemic dysfunction does not correlate with release of cardiac troponin T in isolated rat hearts

Cardiac troponin T (cTnT) is a highly sensitive and specific serum marker of irreversible cardiomyocyte injury. It is not clear whether cTnT also is a suitable marker of subtle, reversible injury. In the present investigation the relationship between cTnT release and function during the first 30 min of reperfusion after 30 min of global ischaemia, was investigated in isolated, retrogradely perfused rat hearts. Left ventricular systolic (LVSP), end-diastolic (LVEDP) and developed (LVDP) pressures, heart rate (HR), and coronary flow (CF) were measured. In one series of experiments (n=7) the kinetics of cTnT release during 30 min of reperfusion was investigated. An early, short-lasting peak of cTnT release appeared after 30 s of reperfusion. Then cTnT release gradually increased with a maximum after 20 min (from 0.08 +/- 0.03 before ischaemia to 2.16 +/- 0.40 ng min-1) (mean +/- SEM). In a second series of experiments (n=52) the relationship between cTnT release and cardiac function was investigated after 20 min of reperfusion. At this time point LVEDP increased (0 to 62 +/- 3 mmHg) and LVDP decreased (84 +/- 2 to 33 +/- 3 mmHg), but without any correlation with cTnT release. cTnT release was positively correlated to LVSP (P < 0.04, r=0.29), and negatively correlated to HR (P < 0.03, r=-0.31). cTnT concentration in the coronary effluent increased in parallel to increasing CF (P < 0.03, r=0.31). In conclusion, during the early reperfusion period there was no consistent correlation between cTnT release and dysfunction after global ischaemia in the isolated rat heart. Release of cTnT and post-ischaemic function appear to provide supplementary information in this particular model.

Computed tomography of the sternum and mediastinum after median sternotomy

BACKGROUND

Computed tomography is used in our hospital to diagnose complications after median sternotomy, but its efficiency is unknown. Nor is the computed tomographic appearance of normal healing of a median sternotomy known. Computed tomography was evaluated for its ability to diagnose mediastinitis and sternal dehiscence, and a reference material of normally healing median sternotomies was created.

METHODS

In a prospective study, 20 patients with a normally healing median sternotomy were examined 1 week, 1 month, 3 months, and 6 months after operation. In a retrospective study, 87 scans from 65 patients that were made because a postoperative complication was suspected were reviewed.

RESULTS

In the prospective study, all patients had clinically uneventful healing. None of the computed tomographic scans showed radiologic signs of healing at 3 months. At 6 months, half of the patients had healed completely. In the retrospective study, 49 scans were performed on suspicion of infection; 7 of them indicated mediastinitis, 2 were false-positive, while mediastinitis was present in a total of 16 of the scans. Thirty-eight scans were made because of sternal pain or suspected dehiscence; after 21 of the scans, recovery was uneventful, and in 11, the definite diagnosis was dehiscence or pseudarthrosis.

CONCLUSIONS

Clinical healing of the sternotomy does not correlate with the computed tomographic image. Computed tomography is not a sensitive tool for diagnosing mediastinitis, and in patients with sternal pain, it adds little information.

Intermittent warm blood cardioplegia does not provide adequate myocardial resuscitation after global ischaemia

OBJECTIVE

Intermittent warm blood cardioplegia is controversial, and many surgeons consider it inadequate for myocardial protection. The purpose of this study was to compare intermittent and continuous warm blood cardioplegia as resuscitation in hearts exposed to global ischaemia.

METHODS

Pigs were put on cardiopulmonary bypass (CPB) and subjected to 30 min of warm, "unprotected", global ischaemia, followed by continuous (n = 7) or intermittent (n = 10, 12 ml/kg every 10 min) warm (34 degrees C) antegrade blood cardioplegia for 45 min (delivery pressure 75-80 mmHg) and weaned from CPB 45 to 60 min later. Indices of left ventricular function were acquired with the conductance catheter technique and pressure-volume loops at baseline and after 90 min of reperfusion.

RESULTS

Cardioplegia was delivered during 17% of the cross-clamp time. Global left ventricular function, evaluated by preload recruitable stroke work (PRSW), was unchanged after continuous cardioplegia; 95 (76-130) (median (quartile interval)) to 91 (90-104) erg/ml x 10(3), but decreased after intermittent cardioplegia; 122 (100-128) to 64 (23-93) erg/ml x 10(3). Two pigs in the intermittent group weaned from CPB, but died before post-bypass measurement. A 95% confidence interval for the difference in post-bypass mean PRSW was estimated as 32 +/- 30 erg/ml x 10(3) (corresponding to P = 0.04 for comparison between treatments). The end-diastolic pressure-volume relation (EDPVR) increased from 0.17 (0.14-0.20) (continuous) and 0.15 (0.12-0.22) (intermittent) mmHg/ml to 0.27 (0.22-0.33) (P = 0.018) and 0.39 (0.25-0.66) (P = 0.005) mmHg/ml, respectively, indicating deterioration in diastolic function. No difference between groups was found in EDPVR, stiffness constant, troponin T release or myocardial water content.

CONCLUSION

Following acute global ischaemia left ventricular global function was, in this model, less preserved using warm intermittent compared to warm continuous cardioplegia.

Warm or cold continuous blood cardioplegia provides similar myocardial protection

BACKGROUND

This study was performed to investigate the effect of temperature of blood cardioplegia on the recovery of postischemic cardiac function.

METHODS

Pigs on cardiopulmonary bypass were subjected to global ischemia (30 minutes), followed by cold (n = 10) or warm (n = 11) continuous antegrade blood cardioplegia (45 minutes) delivered at 55-60 mm Hg.

RESULTS

Global left ventricular function, evaluated by preload recruitable stroke work, decreased with cold cardioplegia from 91 (85-103) [mean (quartile interval)], at baseline, to 73 (55-87) erg x 10(3)/mL postbypass (p = 0.03), but was unchanged after warm cardioplegia; 110 (80-132) to 109 (71-175) erg x 10(3)/mL (p > 0.5). However, the difference between treatment effects was not significant (p = 0.25). Diastolic function, evaluated by end-diastolic pressure-volume relation, deteriorated without any difference between groups. Mean cardioplegic flow was similar between groups. Coronary vascular resistance increased at constant rate during warm cardioplegic delivery, but remained unchanged with cold cardioplegia (p = 0.001 between regression coefficients).

CONCLUSIONS

No significant difference was found in postischemic functional recovery comparing cold and warm continuous blood cardioplegia. Cold cardioplegia is therefore preferred due to added safety of hypothermia.

Endothelial injury and trapping of blood cells in human myocardium following coronary bypass surgery

To investigate the focal myocytic and microvascular injury that develops during the first hour of reperfusion after hypothermic cardioplegic cardiac arrest, and to compare the influence of gentle versus more abrupt reperfusion, serial atrial biopsies were obtained from 14 patients undergoing uneventful coronary bypass surgery. The biopsies were taken before cardioplegia, at the start of reperfusion, and after 20 and 60 min of reperfusion. Transmission electron micrographs of biopsies examined by stereological techniques revealed endothelial injury. Following 20 min reperfusion there was accumulation of both red blood cells (p = 0.03) and polymorphonuclear leucocytes (p = 0.0004) were found. There was also intravascular accumulation of platelets (p = 0.008) and extravasation of red blood cells (p = 0.02), which increased throughout the observation period. If reperfusion was started with a gradual rise in temperature and pressure, the numbers of platelets in the microvessels were lower than following ordinary, abrupt reperfusion (p = 0.06). It is concluded that reperfusion injury is associated with microcirculatory disturbances with trapping of blood cells, changes which may be favourably modified by a gentle reperfusion technique.

Hydrogen peroxide induces endothelial cell atypia and cytoskeleton depolymerization

Reactive oxygen intermediates induce cell injury in a variety of pathophysiological conditions. Human umbilical cord vein endothelial cell (HUVEC) cultures were exposed to 1 or 200 microM H2O2 for 15 min, and observed after 15 min, or 1, 4, 24, or 120 h. Factor VIII and the cytoskeletal proteins vimentin and tubulin were visualized immunocytochemically. Release of lactate dehydrogenase (indices of cell membrane injury) did not increase after H2O2 exposure; nor was cellular expression of factor VIII affected. 200 microM H2O2 induced cell contraction after 15 min which disappeared after 1 and 4 h, but was evident again after 24 h. Immediately after exposure, the filamentous structure of vimentin and tubulin disappeared, but normalized after 1 h. After 120 h, the cytoskeleton filaments were coarsened and disorganized, and an abundance of multinucleated giant cells were observed. Catalase (150 U/ml) abolished all effects of H2O2. One microM H2O2 did not induce any changes in HUVEC. Thus, the present concentrations of H2O2 did not induce cell necrosis or altered expression of factor VIII. Early, reversible cell contraction and depolymerization of cytoskeletal proteins were observed, followed by a delayed contraction and cell atypia after 200 microM H2O2.