Secondary Logo

Journal Logo

Clinical Aspects

MYOCARDIAL STRUCTURAL CHANGES IN LONG-TERM HUMAN SEVERE SEPSIS/SEPTIC SHOCK MAY BE RESPONSIBLE FOR CARDIAC DYSFUNCTION

Rossi, Marcos A.; Celes, Mara R. N.; Prado, Cibele M.; Saggioro, Fabiano P.

Author Information
doi: 10.1097/01.shk.0000235141.05528.47
  • Free

Abstract

INTRODUCTION

Clinical studies have shown that myocardial contractility is reduced in severe sepsis and septic shock in the absence of changes of ventricular preload or afterload (1). This myocardial dysfunction is recognized as an important factor contributing to the high mortality of septic patients (2). Although cardiac function is depressed, it remains controversial whether the myocardial contractile dysfunction is secondary to structural changes. Clinically unrecognized myocardial injury as revealed by elevated levels of cardiac troponins I and T has been reported in critically ill patients (3-7). However, the exact mechanism of the increased troponin levels in patients with sepsis remains unknown. Possible mechanisms are diffuse necrosis, cardiac troponin proteolysis, or leakage of cytoplasmic cardiac troponins with reversible damage to the contractile complex of heart muscle cells. Previous experimental studies have shown either an association between sepsis and structural myocardial changes (8-15), such as tissue and mitochondrial edema and myocyte necrosis, or an impairment of myocardial contractility in the absence of tissue injury (16, 17).

No study, however, has sought to determine the cellular mechanisms directly dependent on sepsis/septic shock that could lead to myocardial structural changes in humans. Previous retrospective study evaluated 71 autopsies of septic patients using morphologic criteria of inclusion: inflammation reaction in two or more organs; acute splenitis; bilateral acute pyelonephritis; fibrin thrombi in the lungs, liver, and glomeruli; and neutrophils sequestration in pulmonary capillaries (18). The authors described a typical infectious myocarditis: 19 cases (27%) of interstitial myocarditis with an exudate mainly composed of neutrophils, sometimes forming microabscesses, 8 cases (11%) of bacterial colonization, and 5 cases (7%) of fiber necrosis. Seventy-one age- and sex-matched cases served as controls. The evaluation was made in formalin-fixed myocardial tissue (with no reference to the specific area of the myocardium) and hematoxylin and eosin-stained sections. Although the authors described a typical bacterial myocarditis, they proposed that these myocardial changes were likely induced by blood-borne mediators and were responsible for the myocardial dysfunction in human sepsis. In the present investigation, we demonstrate greater interstitial cellular infiltration composed of larger and elongated macrophages and TNF-α protein expression in myocytes, interstitial macrophage cell types, and smooth muscle cells and endothelial cells in the vessels; intracardiomyocyte lipid accumulation; scattered foci of disruption of the actin/myosin contractile apparatus; and increased expression for iNOS and nitrotyrosine in myocytes and interstitial macrophage cell types in long-term human septic myocardium as compared with normal and acute pancreatitis control myocardia. These findings give support to an opinion that structural changes could be responsible for sepsis-induced myocardial dysfunction.

MATERIALS AND METHODS

Specimens

Human hearts from eight cases (men and women; age range, 42-78 years) of long-term severe sepsis/septic shock (the patients resided in the ICU an average of 24.3 days, ranging from 8 to 36 days) were selected from autopsies performed at the autopsy service of the University Hospital of the Faculty of Medicine of Ribeirão Preto, University of São Paulo (Table 1). The diagnosis of sepsis was based on the criteria outlined by the ACCP/SCCM Consensus Conference (19). These cases were selected from a great number of autopsies considering (a) the severity of sepsis/septic shock; (b) the long-term residency in the ICU; (c) that no patient had external heart massage, defibrillation, or electrical cardioversion in the previous 7 days; (d) that all patients in the group had no history of heart disease and the postmortem examination revealed their hearts to be without gross deformities of valves, myocardium, and coronary vessels; and (e) that all patients had culture proven sepsis: Gram-positive organisms were isolated in three cases (Staphylococcus aureus in all three cases), Gram-negative organisms were isolated in four cases (Escherichia coli in one case, Pseudomonas aeruginosa and Enterococcus durans in one case, P. aeruginosa in one case, and Acitenobacter sp. in one case) and fungus was isolated in one case (Candida tropicalis). Four were surgical patients and the other four were medical patients. All patients presented acute distress respiratory syndrome, seven died with refractory cardiovascular shock and one with multiorgan failure. At autopsy, all patients presented acute lung injury characterized by lungs heavy, firm, red, congestion, interstitial and intraalveolar edema, alveolar collapse, inflammation, fibrin deposition, and hyaline membranes lining the alveolar walls. Eight human hearts from cases of severe acute necrotizing pancreatitis associated with acute lung injury (men and women; age range, 36-54 years) were selected from autopsies performed at the Death Verification Service of Ribeirão Preto, University of São Paulo (all patients died at home). The acute lung injury was assumed to occur as a consequence of uncontrolled systemic inflammatory response. The diagnosis of acute pancreatitis was based on the presence of extensive necrosis of the pancreatic tissue, affecting acinar and ductal tissues as well as islets of Langerhans. The diagnosis of acute lung injury was based on the observation of the same findings reported for the acute lung injury in the severe sepsis/septic shock group. Three human hearts from cases of accidental death without thoracic injury (men and women; age range, 16-40 years) were obtained from the Medico-Legal Institute, Ribeirão Preto. These latter two groups served as acute pancreatitis control myocardium and normal control myocardium, respectively. The sepsis hearts were obtained within 8 h after death, the acute pancreatitis hearts were obtained within 28 h after death, and the normal control hearts were obtained from 1½ to 24 h after death. The hearts were removed by sectioning the blood vessels near the pericardial sac. The chambers were carefully washed with running water and were examined macroscopically. A cross-section of the entire heart was made perpendicular to the left ventricular axis at the level of the papillary muscles (medioventricular). Myocardial specimens were sampled in the lateral aspect of the left ventricular free wall, halfway between the base and the apex. Transmural blocks of myocardial tissue (4-mm-thick) were excised from the middle portion of the left ventricular free wall, fixed in neutral 10% formalin and processed for conventional light microscopy. Sections (5-μm-thick) were stained with hematoxylin and eosin and phosphotungstic acid-hematoxylin (PTAH) and were examined in a light microscope. For immunohistochemistry, 5-μm-thick sections were placed on silane-coated slides, deparaffinized, washed in phosphate-buffered saline, and then submitted to heat-induced antigen, endogenous peroxidase inhibition, and nonspecific antibody antigen block. Next, the sections were incubated with the primary antibodies. Additional fragments were excised and immediately frozen in isopentane, chilled in liquid nitrogen, and stored accordingly at −80°C for immunohistochemical and histochemical studies.

Table 1
Table 1:
Clinical characteristic of severe sepsis/septic patients

Immunohistochemistry (paraffin sections) and immunofluorescence (frozen sections)

The avidin-biotin-peroxidase complex method and streptavidin-biotin-enzyme complex were used with antibodies against macrophages (monoclonal mouse antihuman macrophages; DAKO (Glostrup, Denmark); CD68, clone KP1; 1:1000), TNF-α (monoclonal mouse anti-TNF-α; Santa Cruz; 1:50), inducible nitric oxide synthase (iNOS) (polyclonal rabbit anti-iNOS; Santa Cruz (CA); 1:25), and nitrotyrosine (polyclonal antinitrotyrosine: Upstate (Lake Placid, NY); 1:100). Omission of the primary antibodies served as negative controls.

Frozen sections (5-μm-thick) were prepared using a cryostat, transferred to silane-coated glass slides, and fixed in ice-cold acetone for immunofluorescence. Immunolabeling was performed using primary antibodies to myosin (mouse monoclonal antiheavy chain cardiac myosin; Abcam (Cambridge, MA); 1:200) and actin (Alexa Fluor 594 phalloidin; Molecular Probes (Eugene, OR); 1:200). The detection system for myosin was fluorescein antimouse IgG (Vector Laboratories (Burlingame, CA); 1:200). Sections were examined using a Leica DMR fluorescence microscope using blue 480/40-nm excitation filter and 527/30-nm suppression filter for visualization of myosin as well as green 515- to 560-nm excitation filter and 590-nm suppression filter for visualization of actin.

Histological estimation for lipid accumulation

For examination of lipid accumulation, 10-μm-thick frozen sections obtained on the cryostat were transferred to glass slides, placed in absolute propylene glycol for 2 min, and stained with oil red O. The stained sections were examined in a Leica DMR microscope.

The lipid accumulation in the heart muscle cells was estimated using a semiquantitative score of 0 to 4 (0 = no fat; 1 = slight; 2 = slight to moderate; 3 = moderate, and 4 = severe). All histological material was coded so that examination could be done blindly by a skilled histopathologist.

Quantitative analysis of macrophages

Macrophages were enumerated and their area was determined using a Leica QWin software (Leica Microsystems Image Solutions, Cambridge, UK) in conjunction with a Leica microscope DMR (Leica, Microsystems GmbH, Wetzlar, Germany), videocamera (Leica Microsystems Ltd, Heerbrugg, Switzerland), and an online computer. The numbers of CD68+ cells were determined using a ×40 objective lens. Measurements were made by a skilled observer blinded to the treatment groups. The fields, measuring 0.089 mm2 each, were chosen at random. At least 25 random fields were counted so that the percentage standard deviation (SD) was <5% for at least five fields. Very large vessels were excluded from the counts.

Statistics

Data were analyzed using a GraphPad Prism statistic program (GraphPad Software Inc, San Diego, CA, USA) for an IBM PC computer. For analysis of macrophages, number and size one-way ANOVA and the Tukey test (for multiple comparisons) were used. Comparison of oil red O staining among the three groups was performed by Kruskal-Wallis and the Dunn tests for multiple comparisons. One value for each case was entered into the analysis. A level of significance of 5% was chosen to denote the difference between means. Unless specified, the data are presented as mean ± SD. Because the sample sizes were small, the SigmaStat program for Windows, version 3.11 (Systat Software, Inc, Richmond, CA, USA), for computing statistical power a posteriori to the collection of data was used.

RESULTS

Description of hematoxylin and eosin sections

Nonspecific myocardial changes were observed in the three groups studied. These included limited areas of elongated myofibers, mildly hypertrophied cardiomyocytes (three cases in septic group, 37.5%, and four cases in acute pancreatitis control group, 50%), mild to moderate interstitial edema, discrete interstitial fibrosis (three cases in septic group, 37.5%, two cases in acute pancreatitis control group, 25%, and one case in normal control group, 33%), and small clusters of "wavy" myocytes with condensed cytoplasm and preserved nuclei.

Quantitative image analysis of macrophages and TNF-α expression

Monoclonal antibody specific for human macrophages CD68, now named macrosialin, was used. Macrosialin is a pan-macrophage molecule, highly expressed by macrophages, dendritic cells, and osteoclasts. The antigen is primarily expressed as a intracytoplasmatic molecule, probably associated with lysosomal granules (20). Macrophages were identified immunohistochemically appearing as elongated cells with a prolonged and extended cytoplasm. In septic hearts, the macrophages appeared larger and more elongated than those found in normal and acute pancreatitis control hearts. There was greater intensity of staining. The macrophages were scattered within the interstitium (perimysium and endomysium) between myocardial cells, throughout the myocardium. In all hearts, they were concentrated around blood vessels.

Figure 1A shows the box and whisker plot for the number of CD68+ macrophages in the left ventricular free wall of normal control, septic, and acute pancreatitis control hearts. The box and whisker plot graph shows the body of the data, being particularly useful for comparing several batches of data. The box ends at the lower and upper fourth (closely related to the interquartile range) and contains a cross bar at the median. The lines from the each end of the box represent the most remote points. Significantly, the number of macrophages is increased in septic hearts as compared with control hearts. This graph also clearly points out that the findings showed a small variability within each group. Figure 1B shows the mean number of macrophages per field of 0.089 mm2 ± SD in septic hearts (25.76 ± 0.260), normal control hearts (1.23 ± 0.26), and acute pancreatitis control hearts (4.22 ± 1.70). In addition, the histomorphometric analysis showed that the mean area size of macrophages ± SD from septic hearts was 96.13 ± 10.42 μm2, 256% higher than the value 37.55 ± 1.36 μm2 determined for the normal control hearts and 184% higher than the value 52.30 ± 5.65 μm2 determined for the acute pancreatitis control hearts (Fig. 1C). This can be clearly seen when the frequency distribution of macrophage size (area) in each group was plotted (Fig. 1D). The differences in the mean values in the treatment groups were greater than would be expected by chance; there is a statistical significant difference. Power of performed ANOVA test with α = 0.05 was 1.0 and 0.9869 for mean number of CD68+ macrophages per microscopic field and mean size of CD68+ macrophages, respectively. Power analysis performed a posteriori is used to assess the probability inherent in the design of a test to reject the null hypothesis. The test used was found a posteriori to have a high power, thus justifying the interpretation of the results.

Fig. 1
Fig. 1:
A, Box and whisker plot for the number of CD68+ macrophages in the left free wall of normal control, septic, and acute pancreatitis control hearts. B, Mean number of CD68+ macrophages per microscopic field of 0.089 mm2 in the left free wall normal control, septic, and acute pancreatitis control hearts. C, Mean size (μm2) of CD68+ macrophages in the left free wall normal control, septic, and acute pancreatitis control hearts. D, The difference of mean size of macrophages in septic myocardium as compared with normal and acute pancreatitis control myocardia can be clearly seen when the percentile frequency distribution of macrophage areas in each group is plotted.

The myocardium of four patients with sepsis/septic shock showed diffuse expression of TNF-α protein localized to myocytes, interstitial macrophage cell types, and endothelial cells and smooth muscle cells in the vessels. This cytokine was focally and faintly expressed in the specimens from two septic patients localized to the same structures. Specimens from two septic hearts expressed TNF-α localized to interstitial macrophage type cells. TNF-α protein was expressed in all specimens from acute myocarditis control hearts localized to interstitial macrophage cell types and was not expressed in any of the normal control heart tissues (Fig. 2).

Fig. 2
Fig. 2:
TNF-α immunostaining in normal control, septic, and acute pancreatitis control myocardia. The right upper panel is representative of diffuse expression of TNF-α staining in septic hearts localized to cardiomyocytes, interstitial macrophage cell types (arrow heads), and endothelial cells and smooth muscle cells in the vessels (arrow) (insert). TNF-α is expressed in interstitial macrophage cell types (arrow heads) in acute pancreatitis control myocardium (left lower panel). TNF-α is not expressed in control myocardium (left upper panel). The right lower panel shows the immunostaining pattern in a sepsis heart using a secondary antibody alone (negative control). (Bar = 50 μm).

Intramyocyte lipid accumulation

Representative images of oil red O staining in each group are shown in Figure 3. The graph in Figure 3 demonstrates semiquantitative oil red O score staining data. The lipid staining is significantly increased in septic hearts as compared with normal controls. Although the amount of lipid in the acute pancreatitis control myocardium appears increased as compared with normal controls, there was no statistically significant difference. The estimated amounts of lipid in septic and acute pancreatitis myocardia were not statistically different.

PTAH procedure and actin and myosin expression

The PTAH procedure could demonstrate that all cardiomyocytes from normal and acute pancreatitis control myocardia were regularly stained showing distinct cross-striations. In contrast, the septic tissue showed partial lack of cross-striations or irregularly disorganized cross-striations within the cardiomyocytes (Fig. 4, upper panels).

Fig. 3
Fig. 3:
The three left panels show representative examples of oil red O staining for normal control, septic, and acute pancreatitis control myocardia. The orange granules at the poles of the myocytes nuclei represent lipofuscin pigment. (Bar = 25 μm). The graph shows the semiquantitative oil red O stain scoring. It clearly shows a significantly increased lipid staining in septic hearts as compared with normal controls. Although the amount of lipid appears increased in pancreatitis control myocardium as compared with normal controls and decreased as compared with septic myocardium, there is no statistically significant difference.

In normal and acute pancreatitis control myocardia, the actin and myosin labeling was regular throughout the cardiac muscle cells. On the other hand, in septic tissue, the localization of actin and myosin was disturbed showing disarrangement of the sarcomeres structure and distinct lack of and myosin labeling; in some of the cells, the myosin and actin filaments are completely lost. These features correspond to diffusely scattered foci of disruption of actin and myosin representing focal lysis of myofilaments (Fig. 4, middle and lower panels). The amount of autofluorescent perinuclearly located pigment lipofuscin (orange fluorescence using blue 480/40-nm excitation filter and 527/30-nm suppression filter) appeared increased in the cardiomyocytes of septic hearts. It is well known that the increase of reactive oxygen species in any intracellular compartment promotes lipofuscin formation representing oxidized/cross-linked protein. However, considering that the accumulation of lipofuscin is age related and the age range of the cases in our groups did not match perfectly, we did not measure the accumulation of lipofuscin in the cardiomyocytes of the three groups studied.

iNOS and nitrotyrosine expression

In the septic myocardia, the expression of iNOS was markedly increased as compared with no or very light expression in cardiomyocytes in normal and acute pancreatitis control myocardia. Practically all cardiomyocytes, interstitial macrophage cell types, and small intramyocardial blood vessels smooth muscle cells stained positively for iNOS. However, foci of cardiac muscle cells, particularly around small intramyocardial blood vessels, showed more intense immunolabeling for iNOS (Fig. 5).

Fig. 4
Fig. 4:
The three upper panels show representative examples of PTAH-stained sections. All cardiomyocytes from normal and acute pancreatitis control myocardia are regularly stained showing distinct cross-striations. The septic myocardium shows partial lack of cross-striations or irregularly disorganized cross-striations within cardiomyocytes. Actin (three lower panels) and myosin (three middle panels) immunolabeling in normal and acute pancreatitis control myocardia is regular throughout the cardiac muscle cells. The septic tissue actin and myosin immunolabeling is disturbed with distinct lack of actin and myosin (arrows) corresponding to actin/myosin disruption. (Bar = 50 μm).

In septic hearts, the expression of nitrotyrosine was markedly increased as compared with no or very light expression in cardiomyocytes of both control hearts. Similar to the distribution of iNOS, practically all myocytes, all interstitial macrophage cell types, and small intramyocardial blood vessels smooth muscle cells stained positively for nitrotyrosine protein by immunolabeling. Also, in the same way as for iNOS, the immunoreactivity for nitrotyrosine was more intense in cardiomyocytes around small intramyocardial blood vessels (Fig. 5).

Fig. 5
Fig. 5:
Representative examples of iNOS (three upper left panels) and nitrotyrosine (three upper right panels) expression in normal control, septic, and acute pancreatitis control hearts. The two lower panels show the immunostaining pattern obtained in a septic heart using a secondary antibody alone (negative control). The immunoreactivity of both iNOS and nitrotyrosine is intense in cardiomyocytes, interstitial macrophage cell types, and small intramyocardial blood vessels smooth muscle cells of septic hearts as compared with no or very light expression in cardiomyocytes in normal and acute pancreatitis control myocardia. Foci of myocytes around small intramyocardial blood vessels show more intense immunolabeling for iNOS and nitrotyrosine. (Bar = 50 μm).

DISCUSSION

There are four major findings in this study. First, septic myocardia showed greater infiltration of macrophages, with larger and more elongated cells, and increased expression of TNF-α protein in comparison with samples taken from normal control hearts and from control hearts taken from patients who died with acute necrotizing pancreatitis. Second, septic myocardia showed intramyocyte lipid accumulation as compared with normal control hearts. Third, the myocardia from normal control and acute pancreatitis control hearts disclosed normal expression of actin and myosin, whereas septic myocardia showed diffusely scattered foci of actin and myosin disruption representing lysis of myofilaments. Fourth, the expression of both iNOS and nitrotyrosine by cardiomyocytes and interstitial macrophage cell types was markedly increased in septic myocardia, distinctively around small intramyocardial vessels, as compared with normal and acute pancreatitis control hearts. Importantly, the findings showed a small variability within each group.

Interstitial macrophages and TNF-α expression

The macrophage is an important effector cell implicated in many cardiac diseases, including atherosclerosis and posttransplantation rejection. This cell has been identified in the interstitial inflammatory infiltrate in diseased hearts, including chronic ischemic heart disease, dilated cardiomyopathy, and myocarditis (21-25). There is growing evidence that implicates macrophages as capable of reducing myocyte function during endotoxemia. In vitro study demonstrated impaired contractility of myocytes treated with supernatant of activated monocytes (26). Another in vitro study showed that the adhesion of monocytes on myocytes via intercellular adhesion molecule-1 reduced myocyte shortening (27). More recently, it has been evidenced that macrophages cause cardiac contractile impairment in experimentally endotoxemia-induced sepsis in mice through TNF-α via both TNF receptors 1 and 2 (TNFR1 and TNFR2) (28). In addition to the increased number of macrophages in the septic myocardium interstitium as compared with control myocardia, both normal and acute pancreatitis, these inflammatory cells showed morphologic features of "activation," appearing larger and more elongated.

In the present study, septic hearts showed expression of TNF-α protein in the myocardium localized to myofibers, interstitial macrophage cell types, and endothelial cells and smooth muscle cells in the vessels. This contrasts with acute pancreatitis control myocardia that expressed TNF-α in interstitial macrophage cell types and normal control myocardia that did not express this cytokine at all. This is in accord with previous findings demonstrating TNF protein expression in myocytes, endothelial cells, smooth muscle cells, and macrophage cell types in adult feline myocardium stimulated by endotoxin in vitro (29). TNF has been claimed to affect myocardium contractile function through disruption of calcium handling that could lead to dysfunctional excitation coupling and consequent systolic and diastolic dysfunction and/or desensitization to calcium mediated by high levels of NO produced by iNOS (30). It is conceivable that increased local release of TNF could be implicated in the reduced cardiac function in sepsis. In addition to dysfunctional myocardium, TNF could be implicated in oxidant stress and direct toxicity (see discussion below on the increase of iNOS expression and evidence for the presence of peroxynitrite.

Intramyocyte lipid accumulation

Intramyocyte lipid accumulation was first reported by Virchow (31) in the myocardium of patients with congestive heart failure, referring to it as "fatty atrophy of the heart." More recently, the same phenomenon has been described as lipotoxicity in various animals with metabolic abnormalities (32). Contractile dysfunction of the heart may ensue. There is evidence in support to this hypothesis in Zucker diabetic fatty rat (33) and in experimentally induced chronic mitral regurgitation in dogs (34). Hearts from these animals showed increased lipid deposition within the cardiomyocytes associated with contractile dysfunction. The septic hearts showed marked intramyocyte accumulation of lipid as revealed by oil red O staining. This probably reflects cardiomyocyte contractile dysfunction. It is known that during heart failure, the myocardium switches from its normally preferred fuel of free fatty acids to glucose due to downregulation of the genes that control free fatty acids oxidation and upregulation of gene expression that regulates glucose oxidation (32). This results in disequilibrium between myocardial fatty acid uptake and oxidation that may lead to myocardial lipid accumulation.

Cross-striations and actin/myosin disruption

The function of the heart alters markedly in sepsis, both physiologically and metabolically. In response to sepsis, the heart dilates and the ejection fractions of both ventricles decrease, with systolic and diastolic dysfunction (2). Although several mediators have been identified, the cause of depressed myocardial performance in septic patients remains unclear. Numerous reports have been suggesting that clinically unrecognized myocardial cell injury/necrosis, as shown by abnormal plasma levels of cardiac troponins, is a marker of cardiac dysfunction (3-7). However, the exact mechanism of these increased troponin levels remains unknown. Possible mechanisms include direct effects of bacterial microorganisms (bacterial myocarditis) and indirect effects of local and circulating mediators (e.g., cytokines or oxygen reactive species) affecting myocardial permeability to macromolecules, endotoxin effects, and ischemic injury due to elevated oxygen consumption or prolonged hypotension (35-38). Our findings clearly showed scattered foci of partial lack of cross-striations or irregularly disorganized cross-striations within cardiomyocytes and scattered foci of disruption of the myofibrillar proteins actin and myosin, representing lysis of myofilaments, in septic hearts. This could account for the increased levels of troponins and the alteration, at least in part, of heart function in sepsis. In the present study, although myocardial function was not examined, the occurrence of hypotension and shock (seven out of eight patients died with refractory cardiovascular shock and one out of eight patients died with multiorgan failure) suggests myocardial depression. In addition, the hallmark of septic shock is a dysfunctional microcirculation that produces regional flow disturbances and abnormal tissue oxygenation, which may cause relative ischemia in various organs, including the heart (37, 39, 40). The focal nature of the lesions suggests that the microcirculation could be involved; that is, the primary site of disease could be at a level capable of causing focal lysis of myofilaments of myocytes in discrete groups.

Increased expression of iNOS and evidence for the presence of peroxynitrite

The increased expression of iNOS by myocardial cells and interstitial macrophage cell types, more intense around small intramyocardial blood vessels, as compared with controls, could imply enhanced production of NO by iNOS, which could contribute to hypotension and myocytolysis (41). Moreover, increased NO production, resulting from cytokine-mediated induction of NO synthase, including local TNF-α protein expression, may exert a negative inotropic effect (42, 43). One could argue that iNOS could be altered in heart samples taken from 1½ to 28 h postmortem. Only weak or focal immunoreactivity was seen in samples taken from normal and hypertensive hearts 6 to 10 h after death (data not shown), whereas diffuse immunolabeling was detected in all hearts from septic patients. This finding is supported by previous observations that show no alteration of iNOS protein expression, as evaluated by immunoblot analysis, in ventricular myocardium taken 6 to 64 h postmortem compared with those obtained immediately after transplantation (44).

The increased presence of nitrotyrosine in both myocardial cells and interstitial macrophage cell types, similarly to iNOS, more intense around small intramyocardial blood vessels, as compared with the immunoreactivity in controls, indicates an increased production of NO and superoxide (O2) that interact to produce peroxynitrite, a powerful oxidizing and nitrating agent that is able to transverse biological membranes. Peroxynitrite nitrates tyrosine residues on proteins to form nitrotyrosine, a stable biomarker of peroxynitrite formation. Previous study demonstrated that myocardial tyrosine nitration occurs in both sepsis and myocarditis and suggested a role for NO-derived peroxynitrite in the myocardial depression associated with these conditions (45). Also, peroxynitrite-mediated tyrosine nitration alters the structure and function of proteins (46). The most abundant proteins for nitration modification within cardiac myocyte are actin and myosin. The observation of scattered foci of actin and myosin filaments disruption in septic hearts supports the idea that tyrosine nitration could potentially decrease myocardial contractility by directly modifying the contractile apparatus; that is, the cardiac myofilaments could be an important target in this setting. Peroxynitrite-related cardiac protein nitration, myofibrillar thinning, and irregular striations patterns have been documented in doxorubicin-treated mice (47). More recently, an in vitro study has demonstrated that the nitration of α-actinin, a protein essential for maintenance of the Z-line and for the integrity of sarcomeres, induces a contractile dysfunction by decreasing the longitudinal transmission of forces between adjacent sarcomeres (48).

In summary, the higher number of macrophages, most of them with morphological features of "activation," and TNF protein expression could favor the reduction of cardiac function in septic hearts. The intramyocyte lipid accumulation in these hearts very likely reflects myocardium ventricular contractile dysfunction. In addition, the increased expression of iNOS and the evidence for the significant presence of peroxynitrite in cardiomyocytes and interstitial macrophage cell types suggest that oxidative damage may play a role in actin/myosin disruption in the hearts of septic patients. Considering the most recent definition of cardiomyopathies of the World Health Organization/International Society and Federation of Cardiology (49), the myocardial changes in long-term human sepsis/septic shock can be classified within the group of specific cardiomyopathies and named "inflammatory cardiomyopathy."

ACKNOWLEDGMENTS

The excellent technical assistance of Monica A. Abreu, Maria Elena Riul, and Lígia B. Santoro is gratefully acknowledged.

REFERENCES

1. Feltes TF, Pignatelli R, Cleinert S, Mariscalco MM: Quantitated left ventricular systolic mechanisms in children with septic shock utilizing noninvasive wall-stress analysis. Crit Care Med 22:1647-1648, 1994.
2. Kumar A, Haery C, Parrillo JE: Myocardial dysfunction in septic shock. Crit Care Clin 16:251-287, 2000.
3. Turner A, Tsamitros M, Bellomo R: Myocardial cell injury in septic shock. Crit Care Med 27:1775-1780, 1999.
4. Arlati S, Brenna S, Prencipe L, Marochi A, Casella GP, Lanzani M, Gandini C: Myocardial necrosis in ICU patients with non-cardiac disease: a prospective study. Intensive Care Med 26:31-37, 2000.
5. ver Elst KM, Spapen HD, Nguyen DN, Garbar C, Huyghens LP, Gorus FK: Cardiac troponins I and T are biological markers of left ventricular dysfunction in septic shock. Clin Chem 46:650-657, 2000.
6. Amman P, Fehr T, Minder EI, Günter C, Bertel O: Elevation of troponin in sepsis and septic shock. Intensive Care Med 27:965-969, 2001.
7. Mehta MJ, Khan IA, Gupta V, Jani K, Gowda RM, Smith PR: Cardiac troponin I predicts myocardial dysfunction and adverse outcome in septic shock. Int J Cardiol 95:13-17, 2004.
8. Coalson JJ, Hinshaw LB, Guenter CA, Berrel EL, Greenfield LJ: Pathophysiologic responses of the subhuman primate in experimental septic shock. Lab Invest 32:561-569, 1975.
9. Mc Donough KH, Lang CH, Spitzer JJ: Depressed function of isolated hearts from hemodynamic septic rats. Circ Shock 12:241-251, 1984.
10. Hersch M, Gnidec AA, Bersten AD, Troster M, Rutledge FS, Sibbald WJ: Histologic and ultrastructural changes in nonpulmonary organs during early hyperdynamic sepsis. Surgery 107:397-410, 1990.
11. Gotloib L, Shostak A, Galdi P, Jaichenko J, Fudin R: Loss of microvascular negative charges accompanied by interstitial edema in septic rats' heart. Circ Shock 36:45-56, 1992.
12. Schlag G, Redl H, van Vuuren CJ, Davies J: Hyperdynamic sepsis in baboons: II. Relation of organ damage to severity of sepsis evaluated by a newly developed morphological scoring system. Circ Shock 38:253-263, 1992.
13. Langenfeld JE, Machiedo GW, Lyons M, Rush BF, Dikdan G, Lysz TW: Correlation between red blood cell deformability and changes in hemodynamic function. Surgery 116:859-867, 1994.
14. Solomon MA, Correa R, Alexander HR, Koev LA, Cobb JP, Kim DK, Roberts WC, Quezado ZM, Scholz TD, Cunnion RE, et alet al: Myocardial energy metabolism and morphology in a canine model of sepsis. Am J Physiol 266:H757-H768, 1994.
15. Allard MF, Hogg JC, Walley KR: Myocardial morphometric changes related to decreased contractility after endotoxin (abstract). Am J Physiol 270:H1446-H1452, 1996.
16. Piper RD, Li FY, Myers ML, Sibbald WJ: Structure-function relationships in the septic rat heart. Am J Respir Crit Care Med 156:1473-1482, 1997.
17. Zhou M, Wang P, Chaudry IH: Cardiac contractility and structure are not significantly compromised even during the late hemodynamic stage of sepsis. Shock 9:352-358, 1998.
18. Fernandes CJ Jr, Iervolino M, Neves RA, Sampaio ELM, Knobel E: Interstitial myocarditis in sepsis. Am J Cardiol 74:958, 1994.
19. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RMH, Sibbald WJ: Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. Chest 101:1644-1655, 1992.
20. Gordon S: Macrophage-restricted molecules: role in differentiation and activation. Immunol Lett 65:5-8, 1999.
21. Levine B, Kalman J, Mayer L, Fillit HM, Packer M: Elevated circulating levels of tumor necrosis factor in severe chronic heart failure. N Engl J Med 323:236-241, 1990.
22. Matsumori A, Yamada T, Suzuki H, Matoba Y, Sasayama S: Increased circulating cytokines in patients with myocarditis and cardiomyopathy. Br Heart J 72:561-566, 1994.
23. Torre-Amione G, Kapadia S, Lee J, Durand JB, Bies RD, Young JB, Mann DL: Tumor necrosis factor-alpha and tumor necrosis factor receptors in the failing human heart. Circulation 93:704-711, 1996.
24. Devaux B, Scholz D, Hirche A, Klovekon WP, Schaper J: Upregulation of cell adhesion molecules and the presence of low grade inflammation in human chronic heart failure. Eur Heart J 18:470-479, 1997.
25. Azzawi M, Kan SW, Hillier V, Yonan N, Hutchinson IV, Hasleton PS: The distribution of cardiac macrophages in myocardial ischemia and cardiomyopathy. Histopathology 46:314-319, 2005.
26. Balligand JL, Ungureanu D, Kelly RA, Kobzik L, Pimental D: Abnormal contractile function due to infection of nitric oxide synthesis in rat cardiac myocytes follows exposure to activated macrophage-conditioned medium. J Clin Invest 91:2314-2319, 1993.
27. Simms MG, Walley KR: Activated macrophages decrease rat cardiac myocyte contractility: importance of ICAM-1-dependent adhesion. Am J Physiol 277:H253-H260, 1999.
28. Tavener SA, Kubes P: Cellular and molecular mechanism underlying LPS-associated myocyte impairment. Am J Physiol Heart Circ Physiol 290:H800-H806, 2006.
29. Kapadia S, Lee J, Torre-Amione G, Birdsall HH, Ma TS, Mann DL: Tumor necrosis factor-α gene and protein expression in adult feline myocardium after endotoxin administration. J Clin Invest 96:1042-1052, 1995.
30. Meldrum DR: Tumor necrosis factor in the heart. Am J Physiol 274:R577-R595, 1998.
31. Virchow R: In: Chance F (ed.): Cellular Pathology: As Based Upon Physiological and Pathological Histology, 2nd ed. London: Churchill, p 325, 1858.
32. Taegtmeyer H, Golfman L, Sharma S, Razeghi P, Van Arstall M: Linking gene expression to function: metabolic flexibility in the normal and diseased heart. Ann N Y Acad Sci 1005:202-213, 2004.
33. Zhou YT, Grayburn P, Karim A, Shimabukuro M, Higa M, Baetens D, Orci L, Unger RH: Lipotoxic heart disease in obese rats: implications for human obesity. Proc Natl Acad Sci U S A 97:1784-1789, 2000.
34. Nemoto S, Razeghi P, Ishiyama M, De Freitas G, Taegtmeyer H, Carabello BA: PPAR-gamma agonist rosiglitazone ameliorates ventricular dysfunction in experimental chronic mitral regurgitation. Am J Physiol Heart Circ Physiol 288:H77-H82, 2005.
35. Parrillo JE: Pathogenetic mechanisms of septic shock. N Engl J Med 328:1471-1477, 1993.
36. Kumar A, Thota V, Dee L, Olson J, Uretz E, Parrillo JE: Tumor necrosis factor alpha and interleukin 1 beta are responsible for in vitro myocardial cell depression induced by human septic shock serum. J Exp Med 183:949-958, 1996.
37. Rossi MA, Santos CS: Sepsis-related microvascular myocardial damage with giant cell inflammation and calcification. Virchows Arch 443:87-92, 2003.
38. Roongsritong C, Warraich I, Bradley C: Common causes of troponin elevations in the absence of acute myocardial infarction: incidence and clinical significance. Chest 125:1877-1884, 2004.
39. Hinshaw LB: Sepsis/septic shock: participation of the microcirculation. An abbreviated review. Crit Care Med 24:1072-1078, 1996.
40. Chagnon F, Bentourkia M, Lecomte R, Lessard M, Lesur O: Endotoxin-induced heart dysfunction in rats: assessment of myocardial perfusion and permeability and the role of fluid resuscitation. Crit Care Med 34:127-133, 2006.
41. Rubanyi GM: Nitric oxide and circulatory shock. Adv Exp Med Biol 454:165-172, 1998.
42. Finkel MS, Oddis CV, Jacob TD, Watkins SC, Hattler BG, Simmons RL: Negative inotropic effects of cytokines on the heart mediated by nitric oxide. Science 257:387-389, 1992.
43. Hare JM, Colucci WS: Role of nitric oxide in the regulation of myocardial function. Prog Cardiovasc Dis 38:155-166, 1995.
44. Thoenes M, Forstermann U, Tracey WR, Bleese NM, Nussler AK, Scholz H, Stein B: Expression of inducible nitric oxide synthase in failing and non-failing human heart. J Mol Cell Cardiol 28:165-169, 1996.
45. Kooy NW, Lewis SJ, Royall JA, Ye YZ, Kelly DR, Beckman JS: Extensive tyrosine nitration in human myocardial inflammation: evidence for the presence of peroxynitrite. Crit Care Med 25:812-819, 1997.
46. Alvarez B, Radi R: Peroxynitrite reactivity with amino acids and proteins. Amino Acids 25:295-311, 2003.
47. Mihm MJ, Yu F, Weinstein DM, Reiser PJ, Bauer JA: Intracellular distribution of peroxynitrite during doxorubicin cardiomyopathy: evidence for selective impairment of myofibrillar creatine kinase. Br J Pharmacol 135:581-588, 2002.
48. Borbély A, Toth A, Edes I, Virag L, Papp JG, Varro A, Paulus WJ, van der Velden J, Stienen GJ, Papp Z: Peroxynitrite-induced alpha-actinin nitration and contractile alterations in isolated human myocardial cells. Cardiovasc Res 67:225-233, 2005.
49. Richardson P, McKenna W, Bristow M, Maisch B, Mautner B, O'Connell J, Olsen E, Goodwin J, Gyrfas I, Martin I, et alet al: Report of the 1995 World Health Organization/International Society and Federation of Cardiology Task Force on the Definition and Classification of Cardiomyopathies. Circulation 93:841-842, 1996.
Keywords:

Sepsis; septic shock; myocardial cell injury; myocardial dysfunction; macrophage; lipid accumulation; actin; myosin; iNOS; nitrotyrosine

©2007The Shock Society