| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Institute for Experimental Medical Research, Ullevaal University Hospital, 2Center for Heart Failure Research, University of Oslo, and 3Department of Cardiology, Ullevaal University Hospital, Oslo, Norway
Submitted 26 August 2004 ; accepted in final form 1 October 2004
 |
ABSTRACT |
|---|
 TOP ABSTRACT METHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
hemodynamic measurements; left ventricular function; heart failure
Heart failure can be categorized as systolic and/or diastolic. Systolic heart failure is characterized by reduced LV contractility, dilatation of the left ventricle, and a reduction in ejection fraction (EF). In postinfarction CHF in both mice and humans, cardiac systolic performance is often impaired, but diastolic dysfunction might also be present. Reduced diastolic performance is associated with reduced relaxation and ventricular filling because of reduced LV compliance leading to elevated LV filling pressure. In postinfarction CHF, the observed elevation of LV filling pressure can be explained by both systolic and diastolic dysfunction. This may be reflected in an elevated LV end-diastolic pressure (LVEDP). The increased LV filling pressure results in elevation of the pressure in the left atrium and in the pulmonary circulation, which may result in distension of the left atrium and pulmonary congestion.
There are several ways of assessing LV function with echocardiographic and LV pressure measurements. LV systolic performance is often measured as EF or as myocardial shortening velocity in humans (10). Although the LV EF reflects myocardial contractility, it has limited value when afterload is abnormal and in valvular disease. LV end-systolic and end-diastolic volumes and dimensions also reflect systolic and contractile performance when afterload is normal. Furthermore, the maximum rate of rise of LV pressure (LVdP/dtmax) is highly sensitive to changes in contractility as seen in CHF, but only at constant LVEDP. LV diastolic performance can be assessed by analysis of LV filling patterns (8, 10). In particular, prolonged deceleration time of early transmitral filling and reduced E/A ratio are useful parameters, because they reflect abnormal LV relaxation and/or diastolic distensibility (15). The maximum rate of decline in LV pressure (LVdP/dtmin), however, is strongly influenced by the pressure at the time of aortic valve closure and is not a good measure of the rate of isovolumetric relaxation. Reduced LV diastolic performance and elevated LVEDP may result in increased left atrial pressure, and dilatation of the left atrium often ensues in humans (10) and larger animals (14). Thus we suggest that left atrial diameter (LAD) might also be a useful parameter in mice.
In the present study, cardiac function and infarct size were assessed by means of echocardiography and invasive LV pressure measurements 3, 5, 7, and 14 days after induction of MI in mice. MI mice with pulmonary congestion were compared with both MI mice without pulmonary congestion and sham-operated mice (Sham).
|
METHODS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Animal model.
MI was induced in 8-wk-old female C57BL/6 mice as previously described (16). Briefly, after being anesthetized with intravenous tail injections of 0.2 ml propofol (10 mg/ml), the animals were tracheotomized and connected to a rodent ventilator (model 874092, B. Braun, Melsungen, Germany) and given a mixture of 1.8–2.2% isoflurane and 
98% oxygen. A thoracotomy was performed in the third intercostal space on the left side, and the left coronary artery was ligated. Sham-operated animals underwent the same procedure except ligation of the artery. At 3, 5, 7, or 14 days after primary surgery, the animals were again anesthetized and intubated before echocardiography, and LV pressure measurements were carried out, as described below. Then the hearts were removed and blotted dry. The right ventricle and atria were removed. In the infarcted hearts, a cut was made longitudinally through the LV free wall from the base to the apex, and, to enable proper unfolding of the infarcted area, one or two further longitudinal cuts were made in the noninfarcted tissue, if necessary. The unfolded endocardia were photographed with a digital camera (Nikon Coolpix 4500). Total endocardial area and infarcted area were then traced (Fig. 1), and infarct size as percentage of total area was calculated by use of ImageJ 1.32j software (Wayne Rasband, National Institute of Mental Health). The LV, infarcted area, right ventricular free wall, and lungs were weighed and normalized to tibial length. A priori we decided that MI mice with lung weight (LW) higher than all Sham animals in the corresponding group should be considered to have pulmonary congestion. No Sham mice had a LW >125% of the mean in their group. A 125% cutoff was close to 4 standard deviations above the mean in the Sham groups. Therefore, MI mice with a LW normalized to tibial length (LW/TL) >125% of average in the corresponding Sham group were considered to have pulmonary congestion subsequent to an MI and were called MIchf. MI mice with a LW/TL <125% compared with Sham were considered not to have a significant pulmonary congestion, and this group was called MI nonfailing (MInf). The average LW/TL values in the MIchf groups were 36–102% higher than in Sham at the various time points. Furthermore, the LW/TL was 78% higher in MIchf compared with MInf at 7 days (P < 0.05), but MInf was not significantly different from Sham.
|
 |
LV circumference and infarcted area were measured in 2D long axis, by freehand tracing of the outer contour of the LV endocardium from LV outflow tract (LVOT) to the LV-left atrial commissure in diastole. Posterior wall shortening velocity (PWSV) and posterior wall relaxation velocity (PWRV) were measured in M-mode by applying a line parallel to the second endocardial contour. Aortic diameter and LAD were measured both in 2D long axis and M-mode.
Doppler measurements.
Doppler recordings were obtained in the left parasternal long axis position. Pulsed Doppler was performed at the level of the LVOT, the right ventricular outflow tract (RVOT), and the mitral valve tips. The Doppler signal from RVOT was almost parallel to the blood flow and therefore measured flow reliably. In LVOT, however, the flow direction deviated 20° from being parallel to the chest wall, and the Doppler beam had an angle of 70° to the chest wall. The deviation between the LVOT flow direction and the ultrasound beam, 50°, was corrected digitally online to avoid underestimation of LVOT flow. Mitral flow velocity was measured at the tips of the mitral valves, with the Doppler beam directed parallel to the flow. Cardiac output (CO) was calculated in LVOT by using the following equation
 |
LV pressure measurements. Immediately after echocardiography was finished, a 1.4-Fr microtip pressure transducer catheter (SPR-671, Millar Instruments, Houston, TX) was introduced into the left ventricle through the right carotid artery for measurements of LV pressures and calculations of its maximal positive and negative first derivatives. Data recorded from 10–15 consecutive beats were analyzed.
Statistics. Data are expressed as group means ± SE, unless indicated otherwise. Comparisons between groups were made using two-way ANOVA and Newman-Keuls post hoc test in Statistica 6.0. Overall changes with time were compared with the value at 3 days in the same group. By convention (12), inter- and intrascorer reliability was expressed as the average difference between the scorers ± 2 SD, corresponding to mean and "limits of agreement" in the Bland-Altman analysis (2). Average absolute difference between the scorers was expressed as means ± SD. Between-scorer difference was tested with a paired t-test. Differences were considered significant for P < 0.05.
|
RESULTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
|
|
|
|
|
|
|
|
Inter- and intrascorer reliability. To assess interscorer reliability, echocardiographic measurements were analyzed by two persons who were blinded for the intervention carried out. Also, one scorer analyzed the measurements twice, on separate days, to evaluate intrascorer reliability. LAD in 2D and M-mode, LVFS, PWSV, and Mit dec were evaluated in a subset of 10 mice. These parameters were chosen because they were considered the most central in describing either increased LV filling pressure, systolic function, or diastolic function. Both intra- and interscorer reliabilities were found acceptable with a mean difference <10% (Table 5). No statistically significant differences were found. Both the intra- and interscorer differences for LAD were as low as 0.00 ± 0.01 and 0.00 ± 0.02 mm, respectively, in both 2D and M-mode.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
Systolic function. We found that LV systolic function was markedly impaired in mice with MIchf. This was evidenced by a depression in LVFS compared with Sham already at 3 days. At 7 days we did not find a significant difference in LVFS between MIchf and MInf. PWSV measures the shortening velocity of viable myocardium in the posterior wall rather than the alterations in LV cavity dimensions measured by LVFS. In our study, mean PWSV in the MIchf groups were significantly lower than in Sham at all time points and also compared with MInf at 7 days. These results are in line with a previous study in postinfarction CHF rats (14).
Subsequent to MI, dilatation of the noninfarcted parts of the LV may follow, which is often accompanied by a shift of the pressure-volume curve to the right. LVDd is therefore related to LV systolic function, and both murine (11) and human (3, 5) studies have shown an increased LVDd in postinfarction CHF. LV dilatation is thought to reflect the increase in LV volume occurring as a compensatory mechanism to restore stroke volume at the early stages. Later it may, however, reflect decompensatory dilatation with hemodynamic deterioration. Gehrmann et al. (7) found that mice subjected to MI had increased apical LV diameters compared with Sham, as early as 24 h postinfarction, but did not investigate at later time points. Gao et al. (6), on the other hand, did not find any change in LVDd after 1 wk, but found an increase after 2.5 wk. Yang et al. (17) found that LVDd had increased after 1 wk and remained stable from 2 wk. The last study is in line with the present study, in which we found an increased LVDd in the MIchf groups compared with Sham as early as at 3 days, as well as a gradual increase until 14 days. Taken together, these studies suggest that LV diameter is increased early after the infarction, gradually dilates during the first 2 wk, and thereafter possibly remains stable.
Catheterization of the LV showed that LVdP/dtmax was reduced in MIchf compared with Sham at every time point measured, and LVdP/dtmax recorded in the MIchf group at 7 days showed a significant difference from both MInf and Sham. This fits with the reduction in PWSV found in MIchf. In agreement with our findings, Gao et al. (6) and Pons et al. (11) both found a reduction in LVdP/dtmax subsequent to MI, but in those studies they did not distinguish between MIchf and MInf.
Diastolic function.
Measuring LV filling patterns by echocardiography is often the preferred method for evaluating LV diastolic function (8, 10, 15). However, estimating the E/A ratio proved difficult in our study, because of fusion of the two waves in the mitral inflow signal in most of the mice. The separation of the E and A waves is heart rate dependent, and in the present study the waves fused at frequencies above 400. On the other hand, our results showed a significant increase in average Mit dec in MIchf compared with Sham at all time points and also a large difference between MIchf and MInf at 7 days. This indicates impaired diastolic function in MIchf, but not in MInf. This parameter has not previously been examined in mice, but from human studies we know that an increase in early deceleration time is associated with reduced LV compliance (10, 15). Another echocardiographic parameter that assesses diastolic function, PWRV, was found to be not different in MIchf compared with MInf and Sham. Thus this parameter might be less sensitive to diastolic dysfunction than Mit dec. Another method used to evaluate relaxation velocities is tissue Doppler. However, this method has not been established in mice, probably because of insufficient time resolution. The invasive diastolic parameter, LVdP/dtmin, was significantly decreased in MIchf compared with MInf, as well as significantly different in the MIchf and Sham groups. This supports the presence of diastolic dysfunction in MIchf.
LVEDP and LAD. The present investigation showed an increase in mean LVEDP in the MIchf groups compared with Sham, but no difference compared with MInf at 7 days. Also, others (9) have found significantly elevated LVEDP in mice with postinfarction heart failure compared with controls but have not distinguished MIchf from MInf. However, LV pressure measurements require cannulation of the right carotid artery, which prevents longitudinal studies. In addition, and especially in mice because of the relatively large catheter dimensions, catheterization may cause damage to the aortic valves and significantly affect cardiac performance. Furthermore, we found no indication that any of the invasive measures could discriminate between the groups any better than the noninvasive echocardiographic parameters, with regard to systolic and diastolic function.
LAD has not been measured in previous studies on mice. However, in humans the finding of an enlarged left atrium is consistent with a diagnosis of diastolic dysfunction (15). As shown in Table 3, mean LAD distinguished the MIchf group from both MInf and Sham and showed a significant increase with time at 7 and 14 days compared with 3 days. This may reflect gradually increasing left atrial pressure, or just a gradual distention of the atrium in response to an elevated but unchanged atrial pressure. When measured in 2D, LAD was larger in MInf than in Sham, despite no change in LW (Fig. 7F). This could be due to elevated atrial pressure in MInf subsequent to a slightly increased (37%) LV filling pressure compared with Sham. However, this rise in LV filling pressure may be insufficient to cause pulmonary congestion.
Echocardiographic method. Echocardiographic descriptions of the murine postinfarction model have been carried out by others recently (1, 6, 11, 17, 18), but the equipment used has often had technical limitations. The depth of interest in the mouse is 0–10 mm, which is too short a distance for ordinary cardiac transducers to be able to focus on the relevant area. To achieve a suitable focus area, high-frequency vessel transducers have been used in previous studies. However, the frame rate was often too slow to truly sample systolic and diastolic frames in 2D. The fully digitized echocardiography system used in the present study had its software modified to allow recordings in 2D at >200 frames per second, which allows sampling of true end-systolic and end-diastolic frames. Also, the use of a 13-MHz transducer especially designed for studies in small rodents allowed for high-resolution imaging combined with a short focus distance (0 mm). These improvements resulted in high-quality recordings in 2D. Furthermore, we were able to analyze our data offline with the opportunity to adjust gain and time scale for optimal resolution. Adjustment of gain and time scale during post hoc analysis may reduce reproducibility compared with conventional analysis of printouts. However, we have shown that our reproducibility is within acceptable limits (Table 5).
Parameters discriminating MIchf, MInf, and Sham. Most investigators agree that there is a gradual transition from asymptomatic compensated heart failure to decompensated symptomatic failure, but it is often difficult to assess the degree of failure in mice with MI. Previous studies on postinfarction mice have not discriminated between failing and nonfailing animals. In the present study, we evaluated an extensive list of parameters that could potentially be used as criteria for postinfarction MIchf. An increase in RVW normalized to body weight has previously been used as an indicator of CHF in the rat (4). Our results showing a significant increase in RVW/TL only at the two later time points could indicate that this parameter discriminates between early and later stages of CHF. However, there was overlap between individual values in the different groups at these time points. Furthermore, RVW will only increase in cases with substantially elevated pressure in the pulmonary circulation. In cases with only modest pulmonary congestion, the pressure in arteria pulmonalis does not increase enough to result in an increase in RVW (3). Thus increased RVW is not a prerequisite for a diagnosis of pulmonary congestion. All MIchf animals had an infarct area >30% of the LV area, whereas none in the MInf group had infarcts that were that large (Fig. 7A). Although measuring infarct area discriminates well on an individual basis, it is, however, terminally invasive. On the other hand, we found that estimating infarct size as percentage of total LV circumference measured noninvasively by 2D echocardiography also discriminated every MIchf animal from the MInf group. All MIchf animals were found to have an infarct >40% of the LV circumference. Although moderately overestimated compared with planimetric measurements of infarct area, the data obtained by the two different methods showed a reasonably good correlation, and therefore one could argue that this might be a very useful parameter for classifying the mice.
Mean LVFS separated both MIchf and MInf from Sham, and there was only one individual result in the MInf group that overlapped with Sham. However, there was an almost complete overlap between individual results in the MIchf and MInf groups, making this parameter unsuited for categorizing the mice. We found that, although LVDd distinguished MIchf, MInf, and Sham well on an average basis, individual values from all three groups overlapped to a certain degree. Sjaastad et al. (14) found that PWSV managed to separate all MIchf from MInf in the rat, but in our investigation there was an overlap between individual measurements in the MIchf and MInf groups. This indicates that PWSV may not be used as a criterion for postinfarction MIchf in mice, even though it can in rats. Of the diastolic parameters, mean Mit dec in MIchf was significantly different from both MInf and Sham and was the parameter best able to separate MIchf from the other groups. However, Mit dec did not separate individual animals in the MIchf group from MInf and Sham. LVdP/dtmin was decreased in MIchf compared with Sham at all time points, and MInf was different from MIchf and Sham. There was, however, overlap between LVdP/dtmin values in all three groups at 7 days. In larger animals, such as rats, LAD has been shown to discriminate between animals with and without heart failure (14). In agreement with this, we found that LAD could on an individual basis completely distinguish MIchf from MInf and Sham. Furthermore, we found this parameter to have a high degree of reliability, judged by a small inter- and intraobserver variability and a high correlation between measurements obtained in 2D and M-mode. For these reasons, we suggest that it is feasible to use this noninvasive measure as a main criterion for postinfarction CHF.
Limitations of the study. This investigation was carried out under specific experimental conditions, including a degree of sedation, certain age of the mice, and certain time from induction of infarction.
In conclusion, our study showed that the systolic parameters LVdP/dtmax, LVFS, and PWSV were decreased in MIchf compared with Sham at all time points investigated. We also found that the diastolic parameters Mit dec and LVdP/dtmin differed significantly in MIchf compared with Sham at all time points. Only LAD and infarct size could distinguish all MIchf animals from both Sham and MInf. We therefore suggest that a significantly increased LAD and infarct size >40% of total LV circumference may serve as major criteria for postinfarction CHF in the mouse.
|
GRANTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMETHODSRESULTSDISCUSSIONGRANTSACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  C. P. Dahl, C. Husberg, L. Gullestad, A. Waehre, J. K. Damas, L. E. Vinge, A. V. Finsen, T. Ueland, G. Florholmen, S. Aakhus, et al. Increased Production of CXCL16 in Experimental and Clinical Heart Failure: A Possible Role in Extracellular Matrix Remodeling Circ Heart Fail, November 1, 2009; 2(6): 624 - 632. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. K. Mork, I. Sjaastad, O. M. Sejersted, and W. E. Louch Slowing of cardiomyocyte Ca2+ release and contraction during heart failure progression in postinfarction mice Am J Physiol Heart Circ Physiol, April 1, 2009; 296(4): H1069 - H1079. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. Krishnamurthy, J. Rajasingh, E. Lambers, G. Qin, D. W. Losordo, and R. Kishore IL-10 Inhibits Inflammation and Attenuates Left Ventricular Remodeling After Myocardial Infarction via Activation of STAT3 and Suppression of HuR Circ. Res., January 30, 2009; 104(2): e9 - e18. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Schumacher, E. Khojeini, and D. Larson ECHO parameters of diastolic dysfunction Perfusion, September 1, 2008; 23(5): 291 - 296. [Abstract] [PDF]
|
|
|
|
|
|
A Platis, Q Yu, D Moore, E. Khojeini, P Tsau, and D. Larson The effect of daily administration of IL-18 on cardiac structure and function Perfusion, July 1, 2008; 23(4): 237 - 242. [Abstract] [PDF]
|
|
|
|
|
|
M. Guazzi, R. Arena, and M. D. Guazzi Evolving changes in lung interstitial fluid content after acute myocardial infarction: mechanisms and pathophysiological correlates Am J Physiol Heart Circ Physiol, March 1, 2008; 294(3): H1357 - H1364. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Subramanian, P. Krishnamurthy, K. Singh, and M. Singh Lack of osteopontin improves cardiac function in streptozotocin-induced diabetic mice Am J Physiol Heart Circ Physiol, January 1, 2007; 292(1): H673 - H683. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P Krishnamurthy, V Subramanian, M Singh, and K Singh Deficiency of {beta}1 integrins results in increased myocardial dysfunction after myocardial infarction Heart, September 1, 2006; 92(9): 1309 - 1315. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K.-O. Larsen, I. Sjaastad, A. Svindland, K. A. Krobert, O. H. Skjonsberg, and G. Christensen Alveolar hypoxia induces left ventricular diastolic dysfunction and reduces phosphorylation of phospholamban in mice Am J Physiol Heart Circ Physiol, August 1, 2006; 291(2): H507 - H516. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. E. Louch, H. K. Mork, J. Sexton, T. A. Stromme, P. Laake, I. Sjaastad, and O. M. Sejersted T-tubule disorganization and reduced synchrony of Ca2+ release in murine cardiomyocytes following myocardial infarction J. Physiol., July 15, 2006; 574(2): 519 - 533. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
, Myocardial infarction (MI) animals with pulmonary congestion (MIchf); 
, nonfailing MI (MInf) animals.
, Sham. *P < 0.05 for MIchf vs. Sham; 
P < 0.05 for MIchf vs. MInf; 
P < 0.05 for MInf vs. Sham.
