|Home | About | Journals | Submit | Contact Us | Français|
Ischemia and reperfusion injury occur in cardiac operations using cardiopulmonary bypass (CPB). Little is known about the immunological and histopathological changes in the atrial tissue under these conditions.
To investigate and compare multiple right atrial biopsy specimens by means of a self-developed pathological and immunohistochemistry panel.
Thirty-six nonselected adult patients (mean age 59±11.6 years, range 34 to 75) who had undergone different types of heart surgery (26 with and 10 without the use of CPB).
Circumscribed necrosis was not found in any of the samples. Contractile bundle necrosis deteriorated only moderately with CPB. The share of hibernated myocardium seemed to increase during CPB, reaching 30% regardless of the basic disease. From the subepicardial toward the subendocardial surface, the amount of contractile proteins decreased continuously. Features similar to those seen with the phenomenon of ‘stunning’, which develops due to acute ischemia, were also noted. The apoptosis index did not exceed 1%. Apoptotic cells were generally randomly spread. It was very characteristic that with the use of CPB neither pro- nor antiapoptotic peptides (Bax, Bcl-2) were seen. In samples taken from patients who underwent surgery performed without the use of CPB both proteins were detected. The occurrence of cellular stress (heat shock protein 70 reaction) was rather variable in the samples.
These investigations should be continued on homogeneous patient populations with the inclusion of proinflammatory cytokine determination.
In general, ventricular tissues of various origins are preferred for histological investigations, even for atrial fibrillation studies. Few studies on atrial tissues, which may serve as a comparison, are available. It is worth mentioning that there is a significant difference between the development of the right and of the left atrium, and that the creatine kinase content of the left atrium has consistently been found to be higher than that of the left atrium (1). It is an old observation that 25% of right atrial myocytes are aberrant – structurally altered – for which significant dilation is primarily responsible. At the same time, it has been shown that various areas of the right atrium might show a totally different histological picture (2). Comparison of positron emission tomography findings with histological patterns of hibernating myocardium, which can now be detected with the periodic acid-Schiff (PAS) method as well, has been a great step forward (3). Ischemic myocardial changes and the role of the various types of necrosis are well known. Acute and chronic ischemic changes, and numerous factors of myocardial protection can be nicely traced histologically. There are reliable tissue signs of hibernation and ischemia. Necrosis, oncosis and apoptosis can be also be distinguished (4). Hematoxylin and eosin (HE) staining of samples to be further investigated is also valuable, and together with phase contrast microscopic examination it has also been found suitable for the detection of early necrosis (5). There are well known methods for histological detection of the contractile proteins and of the signs of progression (6). Immune histological methods have been introduced for the detection of early myocardial injury (7). The role of programmed cell death is in the forefront of interest and has been regarded as important both physiologically and pathologically (8). Simultaneous analysis of apoptosis and necrosis is indispensable, and trichrome staining has gained high appreciation in this field (9). Terminal deoxynucleotidyl transferase-mediated dUTP-X nick end labelling (TUNEL) detection has become a routine method (9,10). The role of apoptosis is indisputable in hibernation, ischemia-reperfusion and the development of congestive heart failure. However, there are certain difficulties and reservations concerning the experimental methods and their interpretation (11). Histological demonstration of contraction band necrosis (CBN) is routine, and the iron hematoxylin method is primarily used for this purpose (12). The PAS staining method is also suitable for showing early stunning; the reaction first appears in the perinuclear area (13). Desmin, which can be verified by immunohistological methods, is located in the area of the Z-disc and plays an important part in alpha-actinin binding (14). During the second window of protection (SWOP), especially following chronic ischemia, explicit positivity of heat shock proteins (HSP) can be seen, and HSP70 can be well standardized. Regional occurrence of various types of HSP can be very diverse (15,16). Their antiapoptotic role might also be decisive (16).
On this basis, we aimed to process right atrial biopsy specimens by means of the above-mentioned methods in a nonrandomized, clinically not preclassified patient population. Fixing samples in formalin and embedding them in paraffin makes the study of archived materials possible. In addition we explored how suitable our pathological and immunohistochemistry panel, previously used to show ventricular and ischemic cardiac changes, is for the assessment of atrial changes. This allowed for histological comparison between ischemic and nonischemic samples, and between findings with and those without the use of cardiopulmonary bypass (CPB) in patients undergoing cardiac surgery.
Thirty-six consecutive (24 male, 10 female) patients participated in this study. The mean age was 59±11.6 years (range 34 to 75). Emergency operation and repeated cardiac surgery were exclusion criteria. Patient and operation characteristics are shown in Table 1.
Intravenous anesthesia was similar in all patients and employed midazolam, alfentanyl, propofol and pipecuronium. All operations except 10 were performed under CPB conditions with normothermia (core temperature 35.6±0.4°C). When CPB was used, the ascending aorta and both caval veins were cannulated separately, and after aortic cross-clamping, myocardial protection was achieved with cold (4°C) antegrade crystalloid cardioplegia (Bretschneider solution) and by topical cooling with ice sludge. Only a cardiac stabilizing device was used during operations carried out without CPB. Coagulation was inhibited with sodium-heparin in all cases. The effect of heparin was antagonized with protamine-sulphate in a 1:1 ratio.
After sternotomy, a transmyocardial tissue block (about 10×5 mm) was taken from the right atrial wall. This biopsy served as a control specimen. Another biopsy of the same size was obtained, proximal to the site of the first one, immediately before aortic cross-clamp release in the case of CPB and at the end of the operation in the case of off-pump surgery. A third tissue block was also made in six patients after CPB cessation. The biopsy samples were immediately fixed in formalin and embedded in paraffin. HE, Heidenhain iron hematoxylin, PAS, and Goldner and Masson trichrome stains were used as routine staining for pathological evaluation.
The following techniques were used:
Manufacturers’ guidelines and recommendations were followed for all methods and reagents. An internal control was used routinely, and in some cases an external, positive control was also applied. Specific, antibody-free samples were used as a negative control. The study was approved by Zala County Hospital Ethics Committee, and verbal and written consent was requested from all patients before enrolment in the study. The investigation complied with the principles of the Declaration of Helsinki.
One sample proved unsuitable for histological investigations because it contained poor material for evaluation. Sixty samples were completely processed.
HE basic staining proved that the samples were suitable for further analysis and that the content of the samples could be visualized. In this study HE staining was suitable for tracing the rupture of intercalated discs. PAS staining showed that regardless of the basic disease or the type of the operation the share of hibernated myocardium in atrial tissue seemed to be increasing during CPB, reaching 30% (Figure 1), and that loss of contractile protein was always accompanied by glycogen accumulation, indicating cardiomyocytes that are alive but nonfunctioning. Heidenhain iron hematoxylin showed that the amount of contractile proteins decreased continuously from the subepicardial toward the subendocardial surface. The Goldner and Masson versions were used to achieve trichrome staining. This method proved to be excellent for the determination of abnormal myocytes and of the amount of contractile proteins. Correspondingly, colliquation myocytolysis was shown almost electively. The method offered outstanding visualization of the rupture of the intercalated discs by illustrating fibrosis and collagen production. Both the amount and the location (focal, segmental or interstitial) of fibrosis were established.
Although normally smooth muscle alpha-actin cannot be visualized in the myocardium, in the present study it was shown in several specimens of the atria obtained after the ischemic period. This suggests acute hypoxia and prolonged hypoperfusion in the atria during CPB (Figure 2).
CBN deteriorated only moderately with CPB (Figure 3). Regarding stress proteins, HSP70 was shown. It was found that the occurrence of cellular stress varied considerably in the samples (Figure 4). The presence of HSP70 serves as an unambiguous sign of SWOP in chronic hypoperfusion. Ischemia-reperfusion and concomitant polymorphonuclear emigration were nicely visualized, especially when trichrome-stained samples were evaluated simultaneously. In the present study, generally, according to strict criteria the amount of apoptosis was less than 1%, and the distribution of apoptosis was generally focal and did not change during any type of cardiac surgery (Figure 5). Proapoptoic oncogen proteins and antiapoptoic proteins were determined and shown by the presence of Bax and Bcl-2, respectively. Unlike findings with ischemic ventricular myocytes (unpublished data), neither of these proteins was found in atrial samples taken during CPB operations. On the contrary, both proteins were detected in samples taken from patients who underwent surgery performed without the use of CPB. Bcl-2 activity was expressed to a greater extent than Bax, which showed weak cytoplasmatic positivity affecting only a few myocytes (Figures 6,,7).7). In the present study, generally these proteins could not be directly attributed to apoptotic cells.
In all of the atrial tissue samples investigated, in general all of these tissue changes and alterations were homogeneously distributed: no subepicardial, subendocardial or midmyocardial predilection was observed. As expected, the amount of contractile proteins in the atrium was small. Tissue changes were rather focal and in some cases segmental. The presence of atrophy and hypertrophy dominated, seemingly independent of the inducing background abnormality. Vessel changes were hardly noted. Dilation, fibre elongation and cardiomyocyte slippage were easily discerned. Tissue changes characteristic of ischemic myocardium were also seen in the atrial samples, even in those from patients who did not suffer from coronary disease. These changes are similar to those found in hibernating myocardial samples. At the same time, in samples taken in the ischemic period of the operations, expressed PAS positivity, strong, occasionally precipitating desmin reaction and early, perinuclear smooth muscle-actin positivity appeared. Persistent damage was manifest in intercalar disc necrosis and in CBN. In atrial tissue samples taken from patients operated on without the use of CPB, Bax and Bcl-2 appeared, especially on the nuclear pole of the myocyte. The methods detailed above are therefore suitable for the detection and vizualization of basic tissue changes and for analysis of atrial myocardial samples. It was concluded that other mechanisms were responsible for the changes found in the present study.
Normal atrial muscle fibres are uniform myocytes with a regular parallel distribution. Delicate connective tissue fields can be seen between the singular myocytes and the small bunches. When explicit alterations are present, big, tortuous bunches can be seen with diverse, irregular myocytes. The average diameter of these is normally around 11.5 μm, increasing to 21.3 μm in pathological conditions. The decrease of contractile proteins is substituted by glycogen accumulation in both animal experiments (17) and human atrial muscle tissue samples (18). The amounts of collagen and elastic fibres are increased significantly in samples taken from abnormal atria.
The nature of tissue changes is the subject of intense debate. Certain authors suggest a degenerative origin (18), while others think that embryonal-fetal dedifferentiation is the proper explanation, and that in this way the cell survives the damaging effect in a hibernated state (2,6). Our observations confirm the suggestion that perinuclear early PAS positivity can be a sign of stunning, which can also develop if the body is cooled down (6). CBN is an early sign of myocytolysis and, although it develops rapidly during hypoxia, it can be provoked by a number of other factors as well (12).
Apoptosis has a characteristic tissue picture that can be shown primarily by careful trichrome impregnation. The normal myocyte nucleus has a tight heterochromatin ring on its periphery, and the remaining euchromatin is spread uniformly in the nuclear matrix. The apoptotic nucleus differs greatly from this shape (6,9,10). When the TUNEL method is used, morphology must be taken into consideration. TUNEL positivity is reported to be 42%, but when the shrunken nuclear morphology was taken into consideration, only 4.2% proved positive (6). Caspase 3 labelling gave results well below 1%. The characteristically shrunken nuclear structure and other features clearly distinguish apoptosis from necrosis. The role of apoptosis is vital in both physiological and pathological processes (19–28). Thus far, data have been available primarily from ventricular samples. Apart from many types of tissue reactions, the characteristic ladder-like gel pattern of the oligonucleosomes is a helpful diagnostic tool (19). We used the reliable TUNEL assay for the detection of apoptosis and we strictly standardized the reaction. If morphological control is applied, other cellular (vessel, inter-stitium) apoptosis can be determined as well.
The role of apoptosis has been accepted in the development of ischemia-reperfusion, chronic ischemia and progressive congestive heart failure (20). At the same time, the real significance and importance of apoptosis is still being debated (21). Although apoptosis is responsible for the physiological elimination of aged cells, numerous factors induce apoptosis. In ischemia, 16, and in congestive heart failure, seven provoking factors have been listed, all presenting mainly with tumour necrosis factor-alpha (TNFα) and Fas effects (23). Under normal circumstances, aging and apoptosis are responsible for the loss of 4% of cells in the left ventricle if 0.1% cardiomyocytes is labelled (24). The role of apoptosis in cell death seems more acceptable in ischemia-reperfusion, although reactive oxygen bursts due to the presence and effects of polymorphonuclear leukocytes must also be supposed (26). This seems to be confirmed by the fact that its degree is decreased by ischemic preconditioning (27,28). Nevertheless it must be taken into consideration that non-apoptotic processes might also result in a positive TUNEL reaction.
The normal cell may possess gross mRNA activity. Because of this phenomenon, it is suggested that DNA repair processes should be investigated simultaneously. According to our investigations the TUNEL method is very sensitive to enzymatic digestion and fixation. The reactions for the simultaneous determination of proaptotic and antiapoptoic proteins were positive only in samples from patients operated on without the use of CPB, resulting in occasionally explicit, but in other cases only perinuclear, reactions. The degree of apoptosis was totally independent of the background disease. It seemed more likely to depend on fibre dilation, slippage and stretching, although there was also focal positivity.
There is no doubt that other factors must also be evaluated because so far tissue reactions to damaging effects have been rather equal. The role of low molecular weight stress proteins is important for more than one reason. They protect the cytoskeletal frame, act against apoptosis, heat and reactive oxygen intermediaries, and possess antiarrhythmogenic properties. Stress proteins also decrease the apoptoic effects of certain agents (chemotherapic agents, Fas L, staurosporin, etopside, etc) (29). Strong reactions have always been observed in the right ventricle during cardiac operations when hypertrophy was also present with reduced oxygen saturation and increased pressure in the background (30). In our study we noticed strong expression in almost each sample and we have regarded it as a part of SWOP. It is a fundamental distinction whether hibernation or tissue changes corresponding to it must be considered as a degenerative process or a dedifferentiated tissue property. Data exist to confirm the theory of both the degenerative-irreversible damage (17,22) and the embryonal-fetal nature (2,6,31). Quantitative and qualitative changes of the contractile proteins are very characteristic and these changes are reflected in alpha-smooth muscle actin positivity (31). In our study this phenomenon was observed in 80%. Slight perinuclear positivity may refer to stunning following acute ischemia. This is of vital importance because if it proves degenerative, urgent revascularization is justified (32). Data have been available in the literature from both animal and human experiments suggesting that in atrial fibrillation hibernation acts as a prelude to apoptosis. Desmin reaction presents in the usual cross-striated pattern, primarily in a desmosomal location, where its presence is intensified. Desmin may play an important part in the rupture of the intercalated discs in hypertrophy (17). In our study we observed mainly its gross, occasionally precipitated presence in hypertrophy. Familial cardiomyopathy developing in the absence of desmin is also well known (33).
Apart from wall tension, fibre elongation and slipping, other factors should also be discussed on the basis of tissue characteristics and data from the literature. Signals resulting in hypertrophy and apoptosis may occasionally lead to contractile dysfunction (34). First is the role of interleukin-6 and TNFα (35), which may even be responsible for the systemic inflammatory response after open heart surgery with CPB (36,37). The cardiomyocyte itself may be responsible for the production of both TNFα and Fas, resulting in a vicious circle, self-intensifying in progressive myocardial abnormalities (38).
Repeated short ischemic episodes may provoke proinflammatory cytokine production and antioxidant enzyme gene induction, all of which may also be responsible for evoking even atrial myocardial abnormalities and hibernation-like tissue changes (39).
In this study we have ascertained that the well known four kinds of basic tissue staining and stress protein, alpha-smooth muscle actin and desmin expression, together with in situ apoptosis studies and with the determination of proaptotic and antiapoptotic effects seem suitable for clarifying the fundamental process patterns in atrial tissue in patients undergoing heart surgery.