INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CONGESTIVE HEART FAILURE (CHF) is an increasingly frequent cause of cardiovascular morbidity and mortality, with ~400,000 new cases reported annually (14, 20). Survival 5 yr after the diagnosis of CHF is poor, ranging as low as 25-35% (8). The loss of functional left ventricular (LV) myocardium after myocardial infarction has replaced hypertension as the primary cause for CHF due to systolic dysfunction (5).

In contrast to the relatively high propensity to develop CHF after myocardial infarction in the human, this syndrome has been difficult to demonstrate in the mouse. In our earlier work and that of others (7, 12, 15, 16, 18), despite clear evidence of depressed systolic and diastolic cardiac function, there was no evidence of a CHF syndrome in young (2-3 mo) mice after coronary artery ligation (CAL). In our hands, mice lived more than 12 mo after CAL with persistently decreased LV systolic function without excess mortality or signs of CHF. To study the biology of heart failure in genetically altered mice, the most clinically relevant etiology was myocardial infarction. Such a model would allow us to reproduce principal aspects of cardiac pathology associated with coronary artery disease and subsequent heart failure, including ischemia, inflammation (3), fibrosis associated with healing of the infarct, asymmetrical loading of the surviving LV and infarct scar (13, 25), compensatory hypertrophy, and ventricular remodeling (15).

The hypothesis of this study was that the use of young mice might have altered the response to myocardial infarction and that older mice might respond in a manner similar to the population in which clinical infarction occurs. Our data show increased mortality, an increase in observable signs of late-stage heart failure, and greater infarct expansion in mice that were at the middle of their life span (14 mo old) at the time of infarction produced by CAL compared with typical 2-mo-old mice. Treatment with the angiotensin-converting enzyme inhibitor captopril increased survival and improved hemodynamic function of older infarcted mice, a finding typical of human CHF. However, captopril did not influence infarct expansion but markedly reduced septal hypertrophy.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Male C57BL/6N mice (Harlan Sprague Dawley; Indianapolis, IN, and Houston, TX) from three age groups were used: 2 mo old (Young mice), 10-12 mo old, and 14 mo old (Older mice). Use of mice was approved by the Institutional Animal Care and Use Committees of the American Association for Accreditation of Laboratory Animal Care-accredited institutions, Lilly Research Laboratories, and Baylor College of Medicine in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals (Revised 1996).

Permanent myocardial infarcts were produced by ligation of the anterior descending branch of the left coronary artery (15, 16). Briefly, mice were anesthetized with intraperitoneal pentobarbital sodium (4 mg/ml, 14 µl/g body wt, Nembutal, Abbott Laboratories; North Chicago, IL), placed in a supine position, and intubated. Mice were ventilated with a mixture of 100% oxygen and room air with a volume-cycled rodent ventilator (100 cycles/min, Harvard Apparatus; Holliston, MA). Ventilator stroke volume was adjusted to fully inflate but not overexpand the lungs. Each paw was taped to a copper electrode for bipolar lead electrocardiographic recordings. A median sternotomy was performed, exposing the anterior surface of the heart. The anterior descending branch of the left main coronary artery (LAD) was ligated at a position ~1 mm from the tip of the normally positioned left auricle with a 8-0 polypropylene monofilament suture (Prolene, Ethicon; Somerville, NJ). The ligature was not removed after placement. Sham operations were created by passing the suture under the coronary artery at the position used for ligation and leaving a short piece of suture in place underneath the artery without ever constricting the artery. After chest closure, mice were recovered in a sternal position, warmed, and provided with 100% oxygen by nose cone. Analgesia was extended with subcutaneous 2.5 mg/kg buprenorphine HCl (Buprenex, Reckitt and Colman Pharmaceuticals; Richmond, VA). Mice recovered overnight in warmed housing with a nebulized oxygen atmosphere and must have survived 24 h after surgery to be included in the study. Mice assigned to control groups did not receive any surgical treatment. All mice were observed daily.

Captopril (Sigma; St. Louis, MO) was dissolved in water and administered by gavage twice daily at 25 mg/kg at each administration, for a total daily dose of 50 mg/kg. The gavage volume was 100 µl. To equalize handling, all mice not receiving captopril were gavaged with an equal amount of water on the same schedule.

Infarct size and the area of myocardium at risk of infarction were measured using 1% Evans blue dye and 1.0% 2,3,5-triphenyltetrazolium chloride (TTC), followed by image analysis as previously reported (15, 16).

Cardiac function was evaluated noninvasively with high-frequency pulsed Doppler ultrasound (6, 7, 21). Mice were anesthetized with a mixture of xylazine (1.4 mg/ml, Xylazine-20, Butler; Columbus, OH), ketamine HCl (42.8 mg/ml, Ketaset, Fort Dodge; Fort Dodge, IA), and acepromazine maleate (1.4 mg/ml, Butler; Columbus, OH) administered intraperitoneally at 0.5 µl of the mixture/g body wt. With mice supine, a 2.0-mm-diameter 10-MHz transducer, mounted at the tip of a stainless steel probe, was placed at the xiphoid process with light pressure. Body fur at the left lower sternal border was clipped lightly, and the skin was wetted with water to improve sound transmission. Optimal Doppler audio flow signals from the outflow tract (aortic root) were obtained by adjusting the position of the transducer on the chest surface and by setting the pulsed Doppler range gate between 4 and 7 mm. The R wave of the electrocardiogram was used as a timing signal and was superimposed on the Doppler spectral display. Multiple sets of signals were obtained from each mouse to allow for variation in heart rate and to assess the consistency of the signals. Doppler signals were acquired for a minimum of 10 s and a maximum of 30 s from the aortic root, yielding up to 150 cardiac cycles for analysis. In the majority of mice, 15-25 beats were analyzed. The pulsed Doppler probes, amplifiers, and range gate circuits were custom made in our own laboratories (6, 7).

An analog spectrum analyzer (Vasculab SP-25A, Medasonics; Mountain View, CA) processed Doppler audio signals for direct viewing and for operator feedback in positioning the probe. Audio signals were simultaneously acquired at 64 kHz and digitized for analysis on a desktop computer programmed and optimized for use with mice (LabVIEW version 4.0, National Instruments; Austin, TX, and VI Engineering; Indianapolis, IN). Each of the treatment groups was followed longitudinally with ultrasound over the time course of the experiment. Data are expressed as a percentage of the individual mouse preprocedure values.

To assess changes in cardiac structure, serial cross sections of the heart were evaluated planimetrically (15). As previously reported (15), hearts were obtained in diastole after cardioplegia was achieved by jugular vein infusion of the following solution (in g/l): 4.0 NaCl, 3.7 KCl, 1.0 NaHCO₃, 2.0 glucose, 3.0 2,3-butanedione monoxime, 3.8 EGTA, and 0.0002 nifedipine (Sigma). Solution pH was adjusted to 7.4. After cardiac standstill was obtained, hearts were excised and rinsed in cold cardioplegia solution. Hearts were fixed for 10 min by retrograde aortic perfusion of 10% neutral buffered zinc-formalin (Z-fix, Anatech; Battle Creek, MI) while immersed in the same fixative. Constant intraventricular fixation pressure of ~16 cmH₂O was maintained by holding the LV drain 16 cm above the base of the heart. Hearts were fixed by further immersion, embedded in paraffin, and sectioned for hematoxylin and eosin or Masson's trichrome staining.

Mass of the LV and interventricular septum and the ventricular expansion ratio (VER) were obtained as previously reported (15).

Statistical analysis of all data was carried out using the SAS and JMP analysis platforms (SAS version 6.12 and JMP version 3.2.2, SAS Institute; Cary, NC). Survival was calculated by Kaplan-Meier life table estimates. Group survivals were compared by a log-rank test for homogeneity. Hemodynamic data from pulsed Doppler ultrasound examinations were analyzed by a mixed-effects repeated-measures ANOVA. The model accounted for effects of treatment group, time of examination, interaction of treatment and time, and heterogeneous variance caused by deaths before the end of a study. The appropriate treatment-by-time interaction contrast and least-squares means were used to assess the magnitude of treatment differences. Anatomic data were analyzed by one-way ANOVA with a contrasting Tukey-Kaplan honest significant differences test or Student's t-test corrected for multiple comparisons as appropriate. Multivariate comparison of interventricular septal mass data and VER was made with a two-sample Hotelling T²-test. All data are reported as means ± SE unless indicated. For all tests, P < 0.05 was used to identify significant differences.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Infarct size 24 h after ligation. Consistent with our prior experience, ligation of the LAD in 2-mo-old C57BL/6 mice (Young mice) placed 43 ± 4% (Fig. 1) of the LV area at risk (AAR) and produced infarctions of 28 ± 4% of the LV when assessed 24 h after CAL (16). Almost 70% (67 ± 4%) of the AAR after CAL could not metabolize the TTC stain at 24 h in Young mice and therefore was considered infarcted. When the LAD was occluded at this location in Older mice, there were no age differences in the LV AAR. Surprisingly, the resulting myocardial infarcts were only 18 ± 2% of the LV, significantly smaller than in the Young mice (P < 0.01 vs. Young mice; Fig. 1). The infarction in Older mice represented only 41 ± 4% of the AAR, significantly less than that observed in Young mice at 24 h (P < 0.001; Fig. 1).

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Infarct size and area at risk after permanent left anterior descending coronary artery occlusion in 3- to 4-mo-old mice (n = 12; Young mice) and 14-mo-old mice (n = 13; Older mice). Hearts were removed at necropsy 24 h after occlusion. Data are means ± SE. *P < 0.011 vs. Older mice for infarct/left ventricle (LV); **P < 0.001 vs. Older mice for infarct/area at risk.

Survival after myocardial infarction. We examined survival in Older mice after CAL (Fig. 2A) compared with young mice. Older mice experienced mortality that was substantially greater than the Young mice (P < 0.001 vs. Older mice). Only 9 of 25 (36%) Older mice survived to the end of the observation period, 56 days after infarction, in contrast to 10 of 10 Young mice (P < 0.02). Ninety-two percent of the sham-operated Older mice survived 8 wk (P < 0.002 vs. untreated infarcted mice; Fig. 2B).

View larger version (23K): [in this window] [in a new window]   Fig. 2.   A: first study of survival to 8 wk after myocardial infarction (MI) at different ages without drug treatment (Young mice: 10 of 10 survived to 8 wk; 10- to 12-mo-old mice: 11 of 18 survived, *P < 0.006 vs. Young mice; Older mice: 3 of 8 survived, *P < 0.047 vs. 10-12 mo old and P < 0.0001 vs. Young mice). B: survival of Older mice to 8 wk after MI and captopril treatment (control: 9 of 9 survived; sham operated: 12 of 13 survived; infarcted captopril treated: 16 of 22 survived, not significant vs. sham-operated mice; infarcted untreated: 9 of 25 survived, *P < 0.02 vs. survival of infarcted Young mice, fP < 0.002 vs. sham-operated mice, P < 0.016 vs. infarcted untreated mice).

Clinical characteristics of mice with heart failure. Over 50% of Older mice developed characteristics after CAL that were different from Young mice. These mice groomed poorly and became increasingly inactive. They showed respiratory distress and clearly visible use of accessory muscles to breathe. Orthostatic apnea was inducible when these mice were placed in dorsal recumbency and was relieved if mice were promptly returned to the sternal position. Rales were occasionally audible but not persistent in individual mice. These findings were also present in those captopril-treated mice that decompensated, as an antecedent to death.

Hemodynamics. Serial studies of blood velocity measurements obtained at the aortic root with pulsed Doppler ultrasound showed progressively greater differences in systolic function when comparing 1) mice that survived to the end of the experiment versus those that died early and 2) untreated versus captopril-treated Older mouse groups (Fig. 3). Two weeks after coronary occlusion, peak aortic velocity in untreated Older mice that survived to the end of the experiment decreased to 85 ± 5% of the preocclusion baseline value. This was a larger decline than that observed in mice that had been receiving captopril for 7 days at the time of the 2-wk measurement (captopril-treated mice: 95 ± 4% of baseline). However, peak aortic velocity in untreated Older mice that died early decreased to 76 ± 2% of the preocclusion baseline value 2 wk after infarction (P < 0.0011 vs. captopril-treated mice). For all groups, mean heart rate was 300-325 beats/min at baseline and was not altered by age, procedure, drug treatment, or outcome.

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Systolic function after MI and captopril treatment obtained by transthoracic pulsed Doppler ultrasound in Older mice. Serial longitudinal measurements of peak aortic blood velocity are shown in indicated cohorts of mice expressed as a percentage of the preinfarction baseline value. All data were obtained through 8 wk after MI except early deaths, which are reported up to 4 wk after infarction. Data are plotted as means ± SE. Preoperative values: control (n = 17), 84.2 ± 2.6 cm/s; sham operated (n = 16), 85.5 ± 2.3 cm/s; infarcted untreated (n = 15), 84.4 ± 2.3 cm/s, and infarcted captopril treated (n = 12), 78.9 ± 3.2 cm/s. 1Significantly different vs. control: P < 0.0296 at 2 wk, P < 0.0122 at 4 wk, and P < 0.0014 at 8 wk. 2Significantly different vs. sham-operated mice: P < 0.0187 at 4 wk and P < 0.0295 at 8 wk. 3Significantly different vs. captopril-treated mice: P < 0.0368 at 8 wk. 4Significantly different vs. control: P < 0.0001 at 2 wk and 4 wk. 5Significantly different vs. sham-operated: P < 0.0002 at 2 wk and P < 0.0001 at 4 wk. 6Significantly different vs. infarcted captopril-treated mice: P < 0.0011 at 2 wk and P < 0.0001 at 4 wk. 7Significantly different vs. surviving infarcted untreated mice: P < 0.0001 at 4 wk.

Light microscopy of myocardial infarcts. Chronic anterior transmural infarcts were observed in the LV of untreated mice that survived to 8 wk. Infarcts included the apex and a major part of the distal ventricular free wall, displaying aneurysmal formation with loss of most viable myocytes characteristic of both old and young mice (16). This appearance was unchanged by captopril in treated mice that survived to 8 wk.

View larger version (21K): [in this window] [in a new window]   Fig. 4.   Ventricular dimensional data from Older (14 mo old) and Young mice (3-4 mo old) 8 wk after MI. A: mass of the interventricular septum. *P < 0.02 vs. all groups. B: ventricular expansion ratio [VER; expressed as a percentage (see METHODS)]. *P < 0.05 vs. Young group sizes: Young sham operated (sham; n = 9), Young MI (n = 12), Older sham (n = 6), Older MI (n = 6), and Older MI + captopril (n = 7). C and D: interventricular septal mass as a function of ventricular expansion. C: Young and untreated Older mice. *P < 0.0099 vs. Young mice. D: Young and Older mice treated with captopril. *P < 0.033 vs. Young group sizes for C and D: Young (n = 11), Older untreated (n = 6), and Older treated with captopril (n = 8). Data are plotted as means ± SE.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

CHF is an increasingly common outcome of myocardial infarction in the United States today (5). In the present study, a heart failure syndrome and increased mortality were found after myocardial infarcts created by CAL in wild-type mice at the middle of their lifespan. This finding is in contrast to the outcome in mice infarcted when 2 mo old (Young mice), in which no heart failure, normal activity, preserved body weight, and very infrequent death were observed for as long as 12 mo after ligation (15). Young mice had persistent and significant cardiac systolic dysfunction as measured by ejection fraction and peak aortic flow velocities (7, 15). The frequent development of CHF in old mice is reminiscent of the higher likelihood of the development after myocardial infarction in aging humans (5, 10) and is substantially different from our experience with young adult mice, those typically used for cardiovascular studies. Additionally, increasing age is associated with increased mortality after myocardial infarction in many studies (1, 11, 22, 23). Importantly, these mice in our hands survive beyond 2 yr, so 14 mo reflects "late middle age," not senescence.

We initially sought to create postinfarction CHF in young wild-type mice but were unable to do so. Male C57BL/6 mice aged 8-10 wk old at the time of coronary occlusion compensated for the substantial injury produced by occlusion of a large coronary artery. Mice of this age had significant decreases in peak aortic velocity and peak diastolic early filling velocity, as shown by pulsed Doppler ultrasound (15). Ejection fractions, determined by ¹⁷⁸Ta ventriculography, were 26.4% at 2 mo after infarction, a 50% reduction from control values (7). Furthermore, infarcted young mice developed none of the clinical characteristics of heart failure even when challenged with 8% dietary salt in the presence of high doses of deoxycorticosterone or when allowed to survive more than 1 yr post-CAL (unpublished data). Therefore, we conclude that myocardial infarction in the young adult mouse is not an optimal model to study the prevention of CHF or accelerated mortality associated with heart failure because the young mouse tolerates extensive myocardial loss without experiencing these clinically relevant endpoints. In contrast, induction of myocardial infarction in the older mouse did result in these outcomes, and they experienced improved survival and reduced septal hypertrophy after angiotensin-converting enzyme inhibition. This latter model is obviously more relevant to the clinical syndrome.

We expected that the infarcts would be larger in the Older mice, consistent with the higher mortality. To our surprise, at 24 h, the infarcted LV mass was significantly smaller than that seen in Young animals. Because the AAR was not different between the two groups, the smaller infarcts could not be explained by any systematic bias in the placing of the ligation or by any age-associated modification in coronary anatomy. Furthermore, at 8 wk, there was substantially greater infarct expansion in the Older hearts. These findings indicate age-related differences in both the acute response of the myocardium to ischemic injury and its subsequent repair in the infarcted myocardial wall.

Even though the initial insult appeared smaller, the greater infarct expansion suggests that wound healing, wall stresses, or other processes important in remodeling may have been modified in the older hearts. Age-related impairments in healing cutaneous and skeletal muscle wounds are found in mice and humans (2). This may be due to a multitude of processes known to be altered by age, including the balance of matrix metalloproteinases and their inhibitors, the entry of inflammatory cells into the infarct zone, the repair activities of the fibroblasts, etc. Similarly, peripheral vasodilatation is also impaired with increasing age, which may have increased afterload and LV wall stress (4, 9, 17). In the present study, Older mice treated with the angiotensin-converting enzyme inhibitor captopril (25 mg · kg¹ · day¹, twice daily, beginning 1 wk after infarction) exhibited significant survival and hemodynamic advantages over untreated mice (Fig. 2B and 3). These benefits occurred without any modification of the infarct expansion but with a decrease in septal hypertrophy. The response to captopril was used to validate the model of postinfarction heart failure and does not allow discrimination between the heart and the periphery as the site of age-related deficit because angiotensin-converting enzyme inhibition has effects at multiple sites (19, 24).

We recognize that this study has a number of limitations. We did not assay neurohumoral activation in these mice, which may have predicted or contributed to the greater mortality in the Older mice. We did not obtain measurements of cardiac loading, which may impact survival and cardiac function. The ligation of a single coronary artery may not duplicate the human condition, where multiple coronary vessels are frequently compromised. Finally, heart rates under anesthesia are slower that those found in the unanesthetized state; however, we have shown that peak aortic flow velocity reveals group differences in systolic function independent of blood pressure (21). Anesthesia effects may have obscured heart rate differences between the age or infarct groups.

In summary, the older male mouse may be a preferred model for studying the outcomes of myocardial infarction after coronary artery ligation, because a number of clinically relevant endpoints may be reproduced.