Exercise training improves cardiac function after ischemia in the isolated, working rat heart D. K. BOWLES, R. P. FARRAR, AND J. W. STARNES Department of Kinesiology, University of Texas at Austin, Austin, Texas 78712 Bowles, D. K., R. P. Farrar, and J. W. Starnes. Exercise training improves cardiac function after ischemia in the isolated, working rat heart. Am. J. Physiol. 263 (Heart Circ. Physiol. 32): H804-H809, 1992.-The aim of this study was to determine whether exercise training produces a myocardium intrinsically more tolerant to ischemic-reperfusioninjury. Male Fischer 344 rats were treadmill trained for 1 l-16 wk at one of the following intensities: LOW (20 m/min, 0% grade, 60 min/ day), moderate (MOD; 30 m/min, 5% grade, 60 min/day) or intensive (INT; 10 bouts of alternating 2-min runs at 16 and 60 m/min, 5% grade). Cardiac function was evaluated both before and after 25 min of global, zero-flow ischemiain the isolated, working heart model. Compared to hearts from sedentary (SED) rats, postischemic cardiac output (CO) and work were significantly higher in all trained groups. Percent recovery of CO (relative to preischemia)was 36.0 t 7.1 in SED and 61.2 & 6.5,68.1 t 9.3, and 73.2 & 5.0 in LOW, MOD, and INT, respectively. Postischemicincreasesin stroke volume with increased preload and cardiac work at high work load were significantly higher in INT comparedwith SED. Coronary flow during initial retrograde reperfusionwassignificantly enhancedwith training and correlated with subsequentrecovery of CO (R2 = 0.613). Furthermore, trained hearts had higher phosphocreatine(P < 0.05) and ATP (P < 0.01) contents after 45 min reperfusion. It is concludedthat exercisetraining results in an intrinsic myocardial adaptation, allowing greater recovery of cardiac pump function after global ischemiain the isolated rat heart. exercise intensity; myocardial dysfunction; no reflow; energy status IT IS KNOWN that chronic exercise training exerts a protective effect against the morbidity and mortality associated with ischemic heart disease (2225, 28). Morris et al. (22) has reported that the incidence of all heart attacks is decreased with physical activity, with the greatest reduction observed for fatal attacks. Thus the survivability of a heart attack is increased with physical activity. Whether this greater survivability is due solely to rheological and peripheral adaptations or whether the myocardium itself becomes more resistant to ischemia is unclear. Studies examining the relationship between exercise training and myocardial ischemic tolerance have been equivocal, with tolerance reported to be improved (6, 7, 29), decreased (19), or unaffected (9, 13, 26). Part of the disparity in the results may stem from the different exercise regimes employed. Most of the investigations have used swim training in rats, which has been shown to decrease infarct size (2l), improve pump function during ischemia (6), and improve functional recovery postcardioplegic arrest (7)) indicating a beneficial training effect. Conversely, hearts from swim-trained animals have also been reported to have decreased time to contracture during total ischemia (19). Swimming, however, elicits dramatically different cardiovascular (12, 31), sympathetic, and hormonal (14) responses compared with treadmill running in the rat. Specifically, EPIDEMIOLOGICALLY,
H804
0363-6135/92
$2.00
Copyright 0 1992
surface swimming results in a redistribution of cardiac output rather than the increase typical of exercise (12). Also, tail weighting and high rat densities in the pool, which are used to increase exercise intensity, can produce a diving reflex and possibly “anoxic training” (31). On this basis, the appropriateness of the swim-trained rat model for mimicking exercise in man can be questioned. Treadmill running in the rat produces hemodynamic changes more similar to those of exercise in humans (12). Hearts from treadmill-trained rats have been reported to demonstrate both an increase and decrease in tolerance during total ischemia. Korge and Mannik (19) reported that half-time to contracture evaluated in excised, totally ischemic hearts was significantly lengthened in rats trained by a moderate, interval-training program, whereas it was shortened by both low-intensity, steady-state and sprint-interval training. Thus intensity of training appears to play an integral role. Also, functional recovery after low-flow ischemia in the isolated, perfused working rat heart is reportedly unaffected by low-intensity treadmill training (26). As the ultimate goal of any prophylactic intervention in ischemia-reperfusion treatment is the resumption of normal contractile and hemodynamic function, it is imperative to examine postischemic contractile function after exercise training. To date, only Paulson et al. (26) have examined hemodynamic recovery after ischemia in treadmill-trained rats, and interpretation of the results is confounded by high glucose perfusion conditions. Thus the purpose of this study was twofold: 1) to determine whether treadmill training in rats can improve recovery of hemodynamic pump function in isolated, working rat hearts after global, normothermic ischemia and 2) to determine whether the extent of myocardial ischemic tolerance is dependent on exercise training intensity. METHODS Animals and training protocols. Male Fischer 344 rats, 8-10 wk old, were obtained from Harlan SpragueDawley (Indianapolis, IN) and housed (3 per cage) in the University of Texas Animal ResourceCenter. All protocols and facilities were approved by the University of Texas Institutional Animal Care and UseCommittee and conform to the American Physiological Society and the Animal Welfare Act guidelines.Rats were randomly divided into four groups:sedentary (SED), low-intensity trained (LOW), moderate-intensity trained (MOD), and interval trained (INT). Some(n = 11) of the SED group were moderately food restricted to match the body weight of the trained animals.Becausethere were no differencesin any hemodynamic parameter at any time, all SED animals were pooled into one SED group. All rats were trained 5 days/wk for 1 l-16 wk between 0700 and 1100h. These hours were during the beginning of the lighted portion of a 12-h light-dark cycle. All training groupswere initially run at 20 m/min, 0% gradefor 5 min/day, the AmericanPhysiological Society
Downloaded from www.physiology.org/journal/ajpheart at Macquarie Univ (137.111.162.020) on February 13, 2019.
EXERCISE
TRAINING
AND MYOCARDIAL
with the intensity and/or duration gradually increasedduring the first few weeks. Final intensities were maintained for at least 5 wk. LOW treadmill training consistedof 20 m/min, 0% grade,60 min/day. The MOD group ran at 30 m/min, 5% grade, 60 min/day. Interval training consistedof 10 repeat bouts of alternating 2-min runs at 16 and 60 m/min. All runs were performed at a 5% gradeand completedwithin 45 min, including a 5-min warm-up run. Maximal oxygen consumption (i702 max) was determined as describedpreviously (10) within 5 days of death. Isolated heart perfusions. Myocardial function wasevaluated using a modified isolated, working heart preparation (23). At least 48 h after the last exercisebout, animalswere anesthetized with pentobarbital sodium (Nembutal; 40 mg/kg ip) and 100 IU heparin injected into the inferior vena cava. A subgroupof LOW (n = 7) was decapitated without anesthesia to determine whether that method of death had any effect on the ischemicreperfusionresponse.No differenceswerefound betweenLOWdecapitated and LOW-anesthetized for any parameter measured; therefore, these groups were pooled for all analyses. Hearts were rapidly excised and placed in ice-cold saline on a tared electronic balancefor determination of grosswet weight. The aortas were secured on a stainless steel cannula of the perfusion apparatusand initially perfusedin a nonrecirculating, retrograde or Langendorff mode at 80 cmH20. Hearts were perfused with a modified Krebs-Henseleit buffer (20) containing (in mM) 10 glucose,0.5 acetate, 1.75 CaCl,, 0.5 EDTA, and 12 IU/l insulin gassedwith 95% 02-5% CO,. During this time extraneous tissue was trimmed from the hearts, weighed, and subtracted from the grossweight to obtain the final wet weight. All subsequentvalues normalizing for heart weight were expressedper gram wet heart weight. The left atrium was cannulated, and a 20-gaugeneedlewasinserted into the left ventricle for determination of left ventricular pressureand its derivative (+dP/dt,,,) using a Gould DTX transducer interfaced with Gould transducer and differential amplifiers. Aortic pressure was determined via an aortic sidearm. Pressuretracings were obtained using a Gould 2200s recorder. Hearts were electrically paced(Phipps & Byrd 611 stimulator) via a platinum wire at or near the sinoatrial node. After 15 min of Langendorff perfusion, hearts were switched to the working heart modeand preischemicheart function was evaluated at a low and high work load. The low work load consisted of 10 cmH,O preload, 80 cmH,O afterload, and a heart rate of 300 beats/min; the high work load consistedof 20 cmH,O preload, 130cmH,O afterload and 420 beats/min. The high work load parameterswere selectedto approach the loading and heart rate conditions of exercisein vivo (12, 14). Intermediate work loadswere accomplishedby increasing preload, heart rate, and afterload, respectively, from low to high work load values at 5-min intervals. Preload (atria1 filling pressure) was measureddirectly and afterload was defined as aortic column (3.18 mm ID) height. Coronary (CF) and aortic (AF) flows were determinedby timed collection of the effluent dripping off the heart and the aortic column overflow, respectively. Cardiac output (CO) was determinedasthe sum of CF and AF. Cardiac external work in the working heart mode is defined as CO x peak aortic systolic pressure (ASP). Before ischemia, hearts were returned to the low work load for 10 min. Global, normothermic ischemia was induced by simultaneouslycross-clampingthe atria1 inflow and aortic outflow lines for 25 min. During ischemia,hearts were enclosedin a sealed, water-jacketed chamber maintained at 37°C and were not paced. After ischemia, hearts were initially reperfused in the Langendorff modefor 15 min at 80 cmH,O before evaluation of postischemicfunction in the working heart mode. Hearts not able to maintain an aortic pressureequal to the aortic column height were assigneda cardiac output value of zero. To separate
H805
ISCHEMIA
the effects of ischemiafrom long-term perfusion, a subgroupof the SED group [nonischemic control (NIC)] was perfusedas above, except the 25 min of ischemiawas replacedby a similar time of normoxic Langendorff perfusion at 80 cmH,O. Enzyme release. Samplesof coronary effluent were collected during O-5 and 5-10 min of reperfusion for determination of lactate dehydrogenase(LDH) content. LDH activity was measured spectrophotometrically at 25°C in 100 mM triethanolamine HCl buffer, pH 7.6, containing (in mM) 150 NADH, 1 EDTA, and 1.5 pyruvate. Enzyme activity is expressedin units where 1 U is that which oxidizes 1 pmol NADH/min. Reperfusion LDH releasewas determined as the product of LDH content and coronary flow rate. Determination of no-reflow areas. After postischemicfunction wasdetermined, someheartswere perfusedwith Monastral blue dye (Sigma Chemical, St. Louis, MO) via the aortic sidearm at a pressureof 80 cmH,O until dye appearedin the coronary effluent. Hearts were then immersedand rinsed in ice-cold saline and subsequentlyimmersion fixed in 10% buffered Formalin for at least 12 h. Fixed hearts were sliced in l-mm sections from apex to base.Four serial sectionsbeginning 2 mm from the apex were photographed with backlighting. Negative imageswere projected with magnification and digitized. Areas of no reflow (lacking dye) were then expressedrelative to the two-dimensionalarea of eachheart slice.Final no-reflow values for eachheart were obtained by averaging the no-reflow areaof three to four slicesfrom each heart. Phosphocreatine and ATP content. Hearts analyzed for energy status were freeze-clampedfollowing postischemicevaluation with aluminum Wollenbergertongs precooledin liquid N2 and then stored at -120°C until analyzed. Frozen portions of the ventricle werepulverized under liquid N2 and extracted with perchloric acid (32). Neutralized extracts were analyzed for phosphocreatine(PC,) and ATP content using standard enzymatic methods (5). Statistical analysis. Data wereanalyzed usinga one-way analysis of variance with post hoc analysis utilizing a Fisher protected least significant difference test for determining differencesbetweenmeans.P < 0.05 wasusedasa limit for statistical significance. All data are expressedas means,t SE. RESULTS
Animal characteristics and Vo2 max at the time of death are presented in Table 1. There were no significant differences in body weight between groups. Heart weights were significantly increased in both LOW and INT groups compared with SED, and in INT vs. MOD. The increased heart weight in the LOW group appeared to be Table 1. Posttraining animal characteris tics Group
Body
Wt, g
Heart
Wt, g
Heart
Wt/Body n-%/g
Wt,
SED
iro2 max, ml-kg-1.min-1
3Olk6 0.85~0.02 2.82kO.04 79.5tl.l (24) (24) (24) (8) LOW 314t3 0.90~0.01* 2.87t0.04 87.9&0.7* (14) (14) (14) (6) MOD 283t9 0.86kO.02 3.07*0.07* 91.9&2.3* (10) (10) (10) (6) INT 293t7 0.93tO.Ol*t 3.17+0.04*3 92.1&0.5*$ (11) (11) (11) (6) Values are means t SE; numbers in parentheses are numbers of rats. VO 2 max9 maximal 0, consumption; SED, sedentary rats; LOW and MOD, low- and moderate-intensity trained rats; INT, interval-trained rats. * Significantly different from SED, P c 0.05; t significantly different from MOD, P c 0.05; “$significantly different from LOW, P c 0.05.
Downloaded from www.physiology.org/journal/ajpheart at Macquarie Univ (137.111.162.020) on February 13, 2019.
H806
EXERCISE
TRAINING
AND MYOCARDIAL
due to increased body weight and not training per se, as only MOD and INT showed significantly larger heart weight-to-body weight ratios compared with SED. . vo 2 max was significantly higher in all trained groups compared with SED. In addition, interval training resulted in significantly higher VO, max compared with LOW. Preischemic cardiac pump function. There were no significant differences in preischemic cardiac pump function due to training at either low or high work load (Table 2). CO, CF, peak ASP, and cardiac power were similar among all groups. Thus cardiac pump performance normalized for tissue mass was similar in all groups before ischemia. Reperfusion coronary flow and LDH release. After 25 min of global ischemia, total coronary flow during the first 10 min of reperfusion CF was significantly increased in all trained groups compared to SED (Table 3). When expressed per gram heart weight, again mean values were higher in all trained groups compared with SED; however, for LOW this difference was not significant (data not shown). A significant positive correlation (R2 = 0.613; P < 0.001) was found among all hearts between reperfusion CF, absolute or normalized, and recovery of cardiac output during reperfusion. At 15 min of reperfusion in the Langendorff mode, absolute CF was significantly higher in all trained groups compared to SED (P < 0.01) and higher in MOD and LOW vs. SED when expressed per gram wet weight (P c 0.05). This increased CF in the hearts of trained animals represents an increased hyperemic response during reperfusion, as mechanical work (heart rate x intraventricular developed pressure) was not different among groups (P > 0.5). When CF at 15 min reperfusion expressed relative to preischemic CF is used as a hyperemic index, all trained groups had significantly increased hyperemia compared with SED (0.87 t 0.04, 1.03 t 0.08, 1.19 t 0.07, and 1.08 t 0.05 for SED, LOW, MOD, and INT, respectively; P < 0.01). Total LDH release during the first 10 min of reperfusion was low and similar among groups (Table 3). No correlation was found between LDH release and functional recovery (see below). Postischemic cardiac pump function. Reperfusion after 25 min of global ischemia resulted in significant contracTable 2. Effects of various intensities of exercise training on preischemic cardiac pump function
co, Group
n
Work Load
ml - min-l -g wet heart wt-’
CF,
ml-min-l -g wet heart wt-’
ASP, mmHg
Work
48.0t0.8 13.OkO.3 91k2 4,386+125 Low 66.1&1.8* 21.7t0.5* 114&2* High 7,582+261* LOW 14 Low 50.0t1.2 13.8k0.3 90&l 4,546+136 High 67.8&2.3* 22.6&0.5* 115&2* 7,953&380* MOD 10 Low 50.6k1.2 14.1t2.0 88t3 4,440&206 High 68.5t2.2* 19.7*1.1* lllt2* 7,538+286* INT 11 Low 47.621.9 11.7t0.4 92t2 4,440+207 High 71.3t2.1* 2 1.4t0.8* 119t3* 8,528t404* Values are means & SE; CO, cardiac output; CF, coronary flow; ASP, peak aortic systolic pressure; work = CO x ASP. * Significantly different from low work load, P < 0.05. SED
24
ISCHEMIA
Table 3. Effects of training on total LDH release and coronary flow during O-10 min of reperfusion after 25 min global &hernia SED
LOW
MOD
INT
LDH, mU/g 2,939+655 3,790+555 3,791+964 2,552+542 wet wt (5) (7) (5) (8) 84.2t5.4 REP CF, 101.8&3.5* 102.5&3.4* 116.3t6.2” ml (15) (12) (9) 03) Values are means & SE; numbers in parentheses are numbers of hearts. LDH, lactate dehydrogenase; REP CF, coronary flow during first 10 min of reperfusion. * Significantly different from SED, P < 0.05.
tile dysfunction in SED (Table 4). Cardiac output and power were significantly depressed compared to preischemic and nonischemic control values (P < 0.05). Hearts from exercise-trained animals showed improved postischemic recovery of cardiac function compared to SED when expressed either relative to preischemic values or in absolute terms (Table 4). At low work load, cardiac output and work were significantly higher in all trained groups. CF was significantly increased in the LOW and MOD compared to SED groups. ASP was depressed in all groups during reperfusion (P < 0.05). At high work load, CO was again higher in all training groups compared with SED. However, CO and ASP were increased over SED only in INT, suggesting a beneficial effect of increased intensity of training. In addition, CO at high work load was increased in INT compared with LOW, and only hearts from the INT group were able to increase CO when work load was increased from low to high (Table 4). Effect of increased preload. The increase in stroke volume (ASV) associated with increasing preload from 10 to 20 cmH20, while maintaining afterload and heart rate constant (80 cmH20 and 300 beats/min, respectively), is shown in Table 4. Exercise training did not affect the magnitude of ASV before ischemia. Ischemia-reperfusion significantly decreased ASV in all groups (P c 0.05); however, during the postischemic period, ASV in INT was significantly higher than SED. Comparison of postischemic +dP/dt,a, values between SED and INT showed no significant differences at either 10 cmH20 (3,000 t 796 vs. 2,880 t 696 mmHg/s, respectively) or 20 cmH20 preload (3,042 t 743 vs. 2,920 t 573 mmHg/s, respectively). Peak left ventricular systolic pressure was not significantly different between groups at 10 cmH20 preload (78 t 3 vs. 85 t 2 mmHg; SED and INT, respectively) but was significantly higher in INT at 20 cmH20 preload (81 t 3 vs. 89 t 2 mmHg; P < 0.05). No reflow. The extent of no-reflow development was evaluated in 25 hearts at the end of reperfusion. Noreflow zones were present in 3 of 9 SED, 1 of 5 MOD, and 1 of 11 LOW hearts (only groups evaluated). Mean areas of no reflow were 16 t 3,17, and 2% for SED, MOD, and LOW, respectively. All hearts exhibiting no-reflow areas had functional recoveries of ~30% (Fig. 1). To estimate whether the lack of no reflow contributed to the improved recovery in the trained P ups, all hea rts with a recovery of
H804
0363-6135/92
$2.00
Copyright 0 1992
surface swimming results in a redistribution of cardiac output rather than the increase typical of exercise (12). Also, tail weighting and high rat densities in the pool, which are used to increase exercise intensity, can produce a diving reflex and possibly “anoxic training” (31). On this basis, the appropriateness of the swim-trained rat model for mimicking exercise in man can be questioned. Treadmill running in the rat produces hemodynamic changes more similar to those of exercise in humans (12). Hearts from treadmill-trained rats have been reported to demonstrate both an increase and decrease in tolerance during total ischemia. Korge and Mannik (19) reported that half-time to contracture evaluated in excised, totally ischemic hearts was significantly lengthened in rats trained by a moderate, interval-training program, whereas it was shortened by both low-intensity, steady-state and sprint-interval training. Thus intensity of training appears to play an integral role. Also, functional recovery after low-flow ischemia in the isolated, perfused working rat heart is reportedly unaffected by low-intensity treadmill training (26). As the ultimate goal of any prophylactic intervention in ischemia-reperfusion treatment is the resumption of normal contractile and hemodynamic function, it is imperative to examine postischemic contractile function after exercise training. To date, only Paulson et al. (26) have examined hemodynamic recovery after ischemia in treadmill-trained rats, and interpretation of the results is confounded by high glucose perfusion conditions. Thus the purpose of this study was twofold: 1) to determine whether treadmill training in rats can improve recovery of hemodynamic pump function in isolated, working rat hearts after global, normothermic ischemia and 2) to determine whether the extent of myocardial ischemic tolerance is dependent on exercise training intensity. METHODS Animals and training protocols. Male Fischer 344 rats, 8-10 wk old, were obtained from Harlan SpragueDawley (Indianapolis, IN) and housed (3 per cage) in the University of Texas Animal ResourceCenter. All protocols and facilities were approved by the University of Texas Institutional Animal Care and UseCommittee and conform to the American Physiological Society and the Animal Welfare Act guidelines.Rats were randomly divided into four groups:sedentary (SED), low-intensity trained (LOW), moderate-intensity trained (MOD), and interval trained (INT). Some(n = 11) of the SED group were moderately food restricted to match the body weight of the trained animals.Becausethere were no differencesin any hemodynamic parameter at any time, all SED animals were pooled into one SED group. All rats were trained 5 days/wk for 1 l-16 wk between 0700 and 1100h. These hours were during the beginning of the lighted portion of a 12-h light-dark cycle. All training groupswere initially run at 20 m/min, 0% gradefor 5 min/day, the AmericanPhysiological Society
Downloaded from www.physiology.org/journal/ajpheart at Macquarie Univ (137.111.162.020) on February 13, 2019.
EXERCISE
TRAINING
AND MYOCARDIAL
with the intensity and/or duration gradually increasedduring the first few weeks. Final intensities were maintained for at least 5 wk. LOW treadmill training consistedof 20 m/min, 0% grade,60 min/day. The MOD group ran at 30 m/min, 5% grade, 60 min/day. Interval training consistedof 10 repeat bouts of alternating 2-min runs at 16 and 60 m/min. All runs were performed at a 5% gradeand completedwithin 45 min, including a 5-min warm-up run. Maximal oxygen consumption (i702 max) was determined as describedpreviously (10) within 5 days of death. Isolated heart perfusions. Myocardial function wasevaluated using a modified isolated, working heart preparation (23). At least 48 h after the last exercisebout, animalswere anesthetized with pentobarbital sodium (Nembutal; 40 mg/kg ip) and 100 IU heparin injected into the inferior vena cava. A subgroupof LOW (n = 7) was decapitated without anesthesia to determine whether that method of death had any effect on the ischemicreperfusionresponse.No differenceswerefound betweenLOWdecapitated and LOW-anesthetized for any parameter measured; therefore, these groups were pooled for all analyses. Hearts were rapidly excised and placed in ice-cold saline on a tared electronic balancefor determination of grosswet weight. The aortas were secured on a stainless steel cannula of the perfusion apparatusand initially perfusedin a nonrecirculating, retrograde or Langendorff mode at 80 cmH20. Hearts were perfused with a modified Krebs-Henseleit buffer (20) containing (in mM) 10 glucose,0.5 acetate, 1.75 CaCl,, 0.5 EDTA, and 12 IU/l insulin gassedwith 95% 02-5% CO,. During this time extraneous tissue was trimmed from the hearts, weighed, and subtracted from the grossweight to obtain the final wet weight. All subsequentvalues normalizing for heart weight were expressedper gram wet heart weight. The left atrium was cannulated, and a 20-gaugeneedlewasinserted into the left ventricle for determination of left ventricular pressureand its derivative (+dP/dt,,,) using a Gould DTX transducer interfaced with Gould transducer and differential amplifiers. Aortic pressure was determined via an aortic sidearm. Pressuretracings were obtained using a Gould 2200s recorder. Hearts were electrically paced(Phipps & Byrd 611 stimulator) via a platinum wire at or near the sinoatrial node. After 15 min of Langendorff perfusion, hearts were switched to the working heart modeand preischemicheart function was evaluated at a low and high work load. The low work load consisted of 10 cmH,O preload, 80 cmH,O afterload, and a heart rate of 300 beats/min; the high work load consistedof 20 cmH,O preload, 130cmH,O afterload and 420 beats/min. The high work load parameterswere selectedto approach the loading and heart rate conditions of exercisein vivo (12, 14). Intermediate work loadswere accomplishedby increasing preload, heart rate, and afterload, respectively, from low to high work load values at 5-min intervals. Preload (atria1 filling pressure) was measureddirectly and afterload was defined as aortic column (3.18 mm ID) height. Coronary (CF) and aortic (AF) flows were determinedby timed collection of the effluent dripping off the heart and the aortic column overflow, respectively. Cardiac output (CO) was determinedasthe sum of CF and AF. Cardiac external work in the working heart mode is defined as CO x peak aortic systolic pressure (ASP). Before ischemia, hearts were returned to the low work load for 10 min. Global, normothermic ischemia was induced by simultaneouslycross-clampingthe atria1 inflow and aortic outflow lines for 25 min. During ischemia,hearts were enclosedin a sealed, water-jacketed chamber maintained at 37°C and were not paced. After ischemia, hearts were initially reperfused in the Langendorff modefor 15 min at 80 cmH,O before evaluation of postischemicfunction in the working heart mode. Hearts not able to maintain an aortic pressureequal to the aortic column height were assigneda cardiac output value of zero. To separate
H805
ISCHEMIA
the effects of ischemiafrom long-term perfusion, a subgroupof the SED group [nonischemic control (NIC)] was perfusedas above, except the 25 min of ischemiawas replacedby a similar time of normoxic Langendorff perfusion at 80 cmH,O. Enzyme release. Samplesof coronary effluent were collected during O-5 and 5-10 min of reperfusion for determination of lactate dehydrogenase(LDH) content. LDH activity was measured spectrophotometrically at 25°C in 100 mM triethanolamine HCl buffer, pH 7.6, containing (in mM) 150 NADH, 1 EDTA, and 1.5 pyruvate. Enzyme activity is expressedin units where 1 U is that which oxidizes 1 pmol NADH/min. Reperfusion LDH releasewas determined as the product of LDH content and coronary flow rate. Determination of no-reflow areas. After postischemicfunction wasdetermined, someheartswere perfusedwith Monastral blue dye (Sigma Chemical, St. Louis, MO) via the aortic sidearm at a pressureof 80 cmH,O until dye appearedin the coronary effluent. Hearts were then immersedand rinsed in ice-cold saline and subsequentlyimmersion fixed in 10% buffered Formalin for at least 12 h. Fixed hearts were sliced in l-mm sections from apex to base.Four serial sectionsbeginning 2 mm from the apex were photographed with backlighting. Negative imageswere projected with magnification and digitized. Areas of no reflow (lacking dye) were then expressedrelative to the two-dimensionalarea of eachheart slice.Final no-reflow values for eachheart were obtained by averaging the no-reflow areaof three to four slicesfrom each heart. Phosphocreatine and ATP content. Hearts analyzed for energy status were freeze-clampedfollowing postischemicevaluation with aluminum Wollenbergertongs precooledin liquid N2 and then stored at -120°C until analyzed. Frozen portions of the ventricle werepulverized under liquid N2 and extracted with perchloric acid (32). Neutralized extracts were analyzed for phosphocreatine(PC,) and ATP content using standard enzymatic methods (5). Statistical analysis. Data wereanalyzed usinga one-way analysis of variance with post hoc analysis utilizing a Fisher protected least significant difference test for determining differencesbetweenmeans.P < 0.05 wasusedasa limit for statistical significance. All data are expressedas means,t SE. RESULTS
Animal characteristics and Vo2 max at the time of death are presented in Table 1. There were no significant differences in body weight between groups. Heart weights were significantly increased in both LOW and INT groups compared with SED, and in INT vs. MOD. The increased heart weight in the LOW group appeared to be Table 1. Posttraining animal characteris tics Group
Body
Wt, g
Heart
Wt, g
Heart
Wt/Body n-%/g
Wt,
SED
iro2 max, ml-kg-1.min-1
3Olk6 0.85~0.02 2.82kO.04 79.5tl.l (24) (24) (24) (8) LOW 314t3 0.90~0.01* 2.87t0.04 87.9&0.7* (14) (14) (14) (6) MOD 283t9 0.86kO.02 3.07*0.07* 91.9&2.3* (10) (10) (10) (6) INT 293t7 0.93tO.Ol*t 3.17+0.04*3 92.1&0.5*$ (11) (11) (11) (6) Values are means t SE; numbers in parentheses are numbers of rats. VO 2 max9 maximal 0, consumption; SED, sedentary rats; LOW and MOD, low- and moderate-intensity trained rats; INT, interval-trained rats. * Significantly different from SED, P c 0.05; t significantly different from MOD, P c 0.05; “$significantly different from LOW, P c 0.05.
Downloaded from www.physiology.org/journal/ajpheart at Macquarie Univ (137.111.162.020) on February 13, 2019.
H806
EXERCISE
TRAINING
AND MYOCARDIAL
due to increased body weight and not training per se, as only MOD and INT showed significantly larger heart weight-to-body weight ratios compared with SED. . vo 2 max was significantly higher in all trained groups compared with SED. In addition, interval training resulted in significantly higher VO, max compared with LOW. Preischemic cardiac pump function. There were no significant differences in preischemic cardiac pump function due to training at either low or high work load (Table 2). CO, CF, peak ASP, and cardiac power were similar among all groups. Thus cardiac pump performance normalized for tissue mass was similar in all groups before ischemia. Reperfusion coronary flow and LDH release. After 25 min of global ischemia, total coronary flow during the first 10 min of reperfusion CF was significantly increased in all trained groups compared to SED (Table 3). When expressed per gram heart weight, again mean values were higher in all trained groups compared with SED; however, for LOW this difference was not significant (data not shown). A significant positive correlation (R2 = 0.613; P < 0.001) was found among all hearts between reperfusion CF, absolute or normalized, and recovery of cardiac output during reperfusion. At 15 min of reperfusion in the Langendorff mode, absolute CF was significantly higher in all trained groups compared to SED (P < 0.01) and higher in MOD and LOW vs. SED when expressed per gram wet weight (P c 0.05). This increased CF in the hearts of trained animals represents an increased hyperemic response during reperfusion, as mechanical work (heart rate x intraventricular developed pressure) was not different among groups (P > 0.5). When CF at 15 min reperfusion expressed relative to preischemic CF is used as a hyperemic index, all trained groups had significantly increased hyperemia compared with SED (0.87 t 0.04, 1.03 t 0.08, 1.19 t 0.07, and 1.08 t 0.05 for SED, LOW, MOD, and INT, respectively; P < 0.01). Total LDH release during the first 10 min of reperfusion was low and similar among groups (Table 3). No correlation was found between LDH release and functional recovery (see below). Postischemic cardiac pump function. Reperfusion after 25 min of global ischemia resulted in significant contracTable 2. Effects of various intensities of exercise training on preischemic cardiac pump function
co, Group
n
Work Load
ml - min-l -g wet heart wt-’
CF,
ml-min-l -g wet heart wt-’
ASP, mmHg
Work
48.0t0.8 13.OkO.3 91k2 4,386+125 Low 66.1&1.8* 21.7t0.5* 114&2* High 7,582+261* LOW 14 Low 50.0t1.2 13.8k0.3 90&l 4,546+136 High 67.8&2.3* 22.6&0.5* 115&2* 7,953&380* MOD 10 Low 50.6k1.2 14.1t2.0 88t3 4,440&206 High 68.5t2.2* 19.7*1.1* lllt2* 7,538+286* INT 11 Low 47.621.9 11.7t0.4 92t2 4,440+207 High 71.3t2.1* 2 1.4t0.8* 119t3* 8,528t404* Values are means & SE; CO, cardiac output; CF, coronary flow; ASP, peak aortic systolic pressure; work = CO x ASP. * Significantly different from low work load, P < 0.05. SED
24
ISCHEMIA
Table 3. Effects of training on total LDH release and coronary flow during O-10 min of reperfusion after 25 min global &hernia SED
LOW
MOD
INT
LDH, mU/g 2,939+655 3,790+555 3,791+964 2,552+542 wet wt (5) (7) (5) (8) 84.2t5.4 REP CF, 101.8&3.5* 102.5&3.4* 116.3t6.2” ml (15) (12) (9) 03) Values are means & SE; numbers in parentheses are numbers of hearts. LDH, lactate dehydrogenase; REP CF, coronary flow during first 10 min of reperfusion. * Significantly different from SED, P < 0.05.
tile dysfunction in SED (Table 4). Cardiac output and power were significantly depressed compared to preischemic and nonischemic control values (P < 0.05). Hearts from exercise-trained animals showed improved postischemic recovery of cardiac function compared to SED when expressed either relative to preischemic values or in absolute terms (Table 4). At low work load, cardiac output and work were significantly higher in all trained groups. CF was significantly increased in the LOW and MOD compared to SED groups. ASP was depressed in all groups during reperfusion (P < 0.05). At high work load, CO was again higher in all training groups compared with SED. However, CO and ASP were increased over SED only in INT, suggesting a beneficial effect of increased intensity of training. In addition, CO at high work load was increased in INT compared with LOW, and only hearts from the INT group were able to increase CO when work load was increased from low to high (Table 4). Effect of increased preload. The increase in stroke volume (ASV) associated with increasing preload from 10 to 20 cmH20, while maintaining afterload and heart rate constant (80 cmH20 and 300 beats/min, respectively), is shown in Table 4. Exercise training did not affect the magnitude of ASV before ischemia. Ischemia-reperfusion significantly decreased ASV in all groups (P c 0.05); however, during the postischemic period, ASV in INT was significantly higher than SED. Comparison of postischemic +dP/dt,a, values between SED and INT showed no significant differences at either 10 cmH20 (3,000 t 796 vs. 2,880 t 696 mmHg/s, respectively) or 20 cmH20 preload (3,042 t 743 vs. 2,920 t 573 mmHg/s, respectively). Peak left ventricular systolic pressure was not significantly different between groups at 10 cmH20 preload (78 t 3 vs. 85 t 2 mmHg; SED and INT, respectively) but was significantly higher in INT at 20 cmH20 preload (81 t 3 vs. 89 t 2 mmHg; P < 0.05). No reflow. The extent of no-reflow development was evaluated in 25 hearts at the end of reperfusion. Noreflow zones were present in 3 of 9 SED, 1 of 5 MOD, and 1 of 11 LOW hearts (only groups evaluated). Mean areas of no reflow were 16 t 3,17, and 2% for SED, MOD, and LOW, respectively. All hearts exhibiting no-reflow areas had functional recoveries of ~30% (Fig. 1). To estimate whether the lack of no reflow contributed to the improved recovery in the trained P ups, all hea rts with a recovery of
