Inconsistent effectiveness of myocardial preservation among cardiac chambers during hypothermic cardioplegia

J THORAC CARDIOVASC SURG 1991;102:684-7

Inconsistent effectiveness of myocardial preservation among cardiac chambers during hypothermic cardioplegia The purpose of tbis study was to examine the selective and differential natures of ischemic injuries among three cardiac chambers (right atrium, right ventricle, and left ventricle) from the viewpoint of

ultrastructural morphometric study. Twenty consecutive adult patients undergoing cardiac operations were studied. The duration of aortic crossclamp time varied from 36 to 142 minutes (mean 83.4 ± 36.4 minutes). Two serial specimens (preischemic and ischemic) were obtained from the right atrium, the right ventricle, and the left ventricle, respectively. A total of 120 biopsy specimens was obtained from these 20 patients. The average mitochondrial surface area of the left ventricle was 0.308 ± 0.062 ILm2 in the preischemic stage and 0.352 ± 0.083 ILm2 in the ischemic stage. This represented a 14.3% increase in mitochondrial surface area after ischemic injury (p < 0.01). The mitochondrial surface area of the right ventricle showed an average increase of 43.7%, from 0.252 ± 0.036 ILm2 in the preischemic stage to 0.362 ± 0.087 ILm2 in the ischemic stage (p < 0.0005). With respect to the mitochondrial surface area of the right atrium, there was an increase of 88.0 %, from 0.217 ± 0.044 ILm2 in the preischemic stage to 0.408 ± 0.084 ILm2 (p < 0.0005). The difference of mitochondrial swelling among three chambers was statistically significant (right atrium versus right ventricle versus left ventricle, p < 0.0005). Moreover, the differences of mitochondrial swelling between any two chambers were also highly significant (right atrium versus right ventricle, p < 0.0005; right ventricle versus left ventricle, p < 0.01; right atrium versus left ventricle, p < 0.0005). In conclusion, our findings suggest that from the viewpoint of ultrastructural morphometric study myocardial injury after an average of 83 minutesof ischemic arrest is poorer in the right chambersof the heart than in the left ventricle, with the right atrium having the poorest preservation.

Ying-Fu Chen, MD, Young-Tso Lin, MD, and Su-Chuan Wu, BS, Taiwan, Republic of China

DesPite a number of experimentaland clinicalstudies on leftventricularpreservation duringcardioplegic arrest, the issue of effectiveness of preservation of the right side of the heart has not beensufficiently addressed. Previously, it wasdemonstrated that atrial noncoronary collateral flow is important in the rapid localwashoutof cardioplegic solution1 and that the use of a singleright atrial cannula for venousdrainage may subject the right sideof the From the Department of Surgery, Kaohsiung Medical College, Kaohsiung, Taiwan, Republic of China. Received for publication Dec. 4, 1989. Accepted for publication July 31, 1990. Address for reprints: Y. F. Chen, MD, Department of Surgery, Kaohsiung Medical College, 100 Shih-Chuan 1st Rd., Kaohsiung, Taiwan, Republic of China.

12/1/24591

684

heart and supraventricularconductionsystemto a significant periodof warm ischemicinjury during cardioplegic arrest.2, 3 As a consequence, some clinicalreports" 5 suggested that protectionof the right side of the heart may be far from ideal. Nevertheless, such ischemic injury of right-sided cardiac chambers is difficult to diagnose and quantitate perioperatively because the standard 12-lead electrocardiogram is unlikelyto showpathologic Q waves with selective right-sidedchambers alone.s Furthermore, becausethe right-sided cardiac chamberscontributeonly a small fraction of the total creatine kinase isoenzyme (CK-MB) normallyreleasedafter cardiac operation, any CK-MB coming from the ischemic right-sided cardiac chambers is often masked.' Technetium pyrophosphate extraction kinetics may provide a means of recognizing the real-time myocardial injury." However, this method also has a lack of selectivity in differentiating injuries

Volume 102 Number 5 November 1991

among the cardiac chambers. For these reasons investigational attempts to elucidate the ischemic injury of the right-sided cardiac chambers should include examination of these heart chambers. However, most earlier attempts to quantitate ventricular functional response to ischemia have been less precise," On the other hand, biochemical determinants of indices of ischemic injury of various cardiac chambers, such as adenine nucleotide concentrations, are hampered in human heart studies by the fact that a large amount of tissue is needed.f In contrast, ultrastructural studies to investigate the ischemic injury in cardiac operations require only small tissue samples. The purpose ofthis study was to examine the selective and differential natures of ischemic injuries among three cardiac chambers from the viewpoint of ultrastructural morphometric study.

Patients and methods Twenty consecutiveadult patients undergoing cardiac operations werestudied. Ages ranged from 19to 63 years (mean 39.1 years). There were 11 male and nine female patients (Table I). Cardiopulmonary bypass was accomplished by double venous cannulation with snares placed on the superior and inferior venae cavae..lmmediately after aortic crossclamping, blood cardioplegic solution, 15 ml/kg body weight, was infused into the aortic root. Additional cardioplegicsolution (7.5 ml/kg) was administered every 30 minutes or if electromechanical activity returned. The composition of blood cardioplegic solutions used was as reported previously." Moderate systemic hypothermia (mean 28.3° ± 1.30 C) was induced with the heat exchanger of the extracorporeal circulation. External topical hypothermia was induced by pouring ice slush on the heart. The duration of aortic crossclamp time varied from 36 to 142 minutes (mean 83.4 ± 36.4 minutes). Cardiopulmonary bypass time ranged from 60 to 302 minutes (mean 136.4 ± 59.5 minutes). Two serial specimens (preischemic and ischemic) were obtained from the right atrium (RA), right ventricle (RV), and left ventricle (LV). The first specimen was taken before aortic crossclamping as a control sample; the second specimen was obtained at the end of ischemia. The right and left ventricular musclebiopsysamples were obtained by direct puncture of right ventricular free wall and the left ventricular apex with a disposable Travenol biopsy needle (Tru-Cut; Baxter Healthcare Corp., Deerfield, IIl.). A total of 120 biopsy specimens was obtained from these 20 patients. Each specimen was fixed for 1'/2 hours in 2% paraformaldehyde, 2.5% glutaraldehyde in cacodylate buffer solution, 0.1 rnmol/L. The specimen was washed several times with cold phosphate buffer solution, 0.1 mmol/L, for 1'/2 hours and then dehydrated with graded ethyl alcohol. Subsequently, the specimen was embedded in epoxy resin and sectioned after dealcoholization with propylene oxide. Finally, it was stained with uranyl acetate and lead citrate and examined with an electron microscope (Hitachi H-500; Hitachi Ltd., Tokyo, Japan). Stereologic morphometric studies. For each of 120 biopsy specimens, 12 electron micrographs were obtained. The initial magnificationof the electron micrograph was X 10,000. Forty to 50 randomly selected mitochondria from each electron micro-

Myocardial preservation

685

Table I. Summary of clinical and operative data No. of patients Valvular heart disease Congenital heart disease Coronary arterydisease Marfan'ssyndrome Male/female Age(yr) Aortic crossclamp time (min) CPB time (min) Systemic hypothermia (0 C)

20 16

2 1 1 11/9 39.1 ± 83.4 ± 136.4 ± 28.3 ±

13.0* (19-63) 36.4* (36-142) 59.5* (60-302) 1.3* (25-30)

CPB, Cardiopulmonary bypass. 'Data are mean :± standard deviation, with range in parentheses.

graph were measured for surface area. Thus 500 mitochondria in each biopsy were assessed. The area of a mitochondrion was traced with a Houston Hipad digitizer (Houston Instrument, Austin, Tex.). The mitochondrial surface area of each electron micrograph was measured with the use of an Apple lIe computer with the Bioquant Image Analysis System (R & M Biometrics, Inc., Nashville, Tenn). To quantitate and compare degrees of mitochondrial swelling after maximal ischemia, we compared the relative changes in size of mitochondria in two serial biopsy specimens of the RA, RV, and LV. Then the severity of mitochondrial swellingof the RA, RV, and LV after maximal ischemia was compared so that the effectiveness of myocardial preservation could be assessed. Statistical methods. Paired data and preischemic and ischemicvalues of calculated mitochondrial surface area within each chamber were analyzed by paired Student's t test with significancedetermined at a probability ofless than 0.05. One-way analysis of variance (ANOVA) was performed to compare the mean percent of mitochondrial swelling among the three cardiac chambers, followed by Tukey's multiple comparisons to determine if ANOVA P value was less than 0.05. Differences were considered significant at a p value of less than 0.05.

Results By stereologic morphometric measurement the average mitochondrial surface area of the LV was 0.308 ± 0.062 ~m2 in the preischemic stage and 0.352 ± 0.083 ~m2 in the ischemic stage. This represented only minor swelling, a 14.3% increase in the mitochondrial surface area after ischemic injury. However, the difference was statistically significant (p < 0.01) (Table II). The mitochondrial surface area of the R V showed an average increase of 43.7%, from 0.252 ± 0.036 ~m2 in the preischemic stage to 0.362 ± 0.087 ~m2 in the ischemic stage. The difference was highly significant (p < 0.0005). With respect to the mitochondrial surface area of the RA, there was an increase of 88.0%, from 0.217 ± 0.044 ~m2 in the preischemic stage to 0.408 ± 0.084 ~m2. The difference was also highly significant (p < 0.0005). The difference of mitochondrial swelling among three chambers is statistically significant (RA versus R V ver-

The Journal of Thoracic and Cardiovascular Surgery

6 8 6 Chen, Lin, Wu

Table II. Mitochondrial surface area" calculatedfrom RA, RV, and LV muscle cells

Patient 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Mean

SD

Aortic crosse/amp time (min) 71 82 128 130 82 126 64 76 50 43 37 85 51 142 82 79 36 38 134 131 83.4 36.4

RV

RA

LV

Preischemia

Ischemia

Preischemia

Ischemia

Preischemia

Ischemia

0.160 0.209 0.223 0.189 0.157 0.205 0.284 0.183 0.272 0.251 0.171 0.200 0.253 0.291 0.252 0.272 0.157 0.235 0.185 0.185 0.217t 0.044

0.359 0.368 0.370 0.470 0.280 0.329 0.509 0.452 0.482 0.505 0.254 0.381 0.500 0.515 0.472 0.374 0.409 0.379 0.273 0.484 0.408t 0.084

0.260 0.228 0.326 0.208 0.240 0.249 0.306 0.218 0.291 0.260 0.261 0.218 0.293 0.226 0.257 0.203 0.288 0.277 0.246 0.191 0.252:1: 0.036

0.322 0.281 0.374 0.361 0.279 0.429 0.396 0.329 0.408 0.543 0.252 0.417 0.545 0.303 0.461 0.282 0.278 0.394 0.314 0.271 0.362:1: 0.087

0.301 0.241 0.329 0.282 0.426 0.290 0.264 0.255 0.279 0.232 0.323 0.331 0.385 0.244 0.396 0.433 0.338 0.292 0.262 0.252 0.308§ 0.062

0.339 0.287 0.308 0.265 0.407 0.429 0.365 0.328 0.293 0.319 0.352 0.543 0.531 0.276 0.454 0.361 0.345 0.258 0.301 0.283 0.352§ 0.083

SD. Standard deviation. • Area given in square micrometers. tp < 0.0005, difference between preischemic and ischemic values of RA. < 0.0005, difference between preischemic and ischemic values of RV. §p < 0.01, difference between preischemic and ischemic values of LV.

*p

Table III. Average mitochondrial surface area calculated from RA, RV, and LV muscle cells Location of biopsy samples

Preischemia"

Ischemia *

RA

0.217 ± 0.044

0.408 ± 0.084

88.0

RV

0.252 ± 0.036

0.362 ± 0.087

43.7

% Difference'[ ]:1: ]§

LV

0.308 ± 0.062

0.352 ± 0.083

:I:

14.3

'Values (square micrometers) shown as mean ± standard deviation. tDifference between preischemic and ischemic values for RA, RV, or LV. :j:p < 0.0005. §p < 0.01.

sus LV, P < 0.0005). Moreover, the differences of mitochondrial swelling between any two chambers are also highlysignificant (RA versus RV,p < 0.0005; RV versus LV, p < 0.01; RA versus LV, p < 0.0005) (Table III). Discussion

The ultrastructural changes of the mitochondria previously have been shown to be one of the earliestsigns of

ischemic injury.'? Assessing the ischemic injury of the mitochondria can clarify whether the effectiveness of myocardial preservation among three cardiac chambers during cardioplegic arrest is uneven. Myocardial oxygen reserves are rapidly depleted during cardioplegic arrest and energydemand cannot be met with anaerobic glycolysis alone. I I The accumulation of the end products of anaerobic glycolysis results in cellularacidosis that inhibits furtheranaerobic glycolysis. 12 Limitation ofanaerobic adenosine triphosphate production leads to inhibition of cellularprocesses regulating cell volume (sodium-potassium adenosine triphosphatase pump) and increases cell membrane permeability. Therefore inability to regulate cellvolume may be an important factor in the pathogenesis of irreversible cellinjury.I 3 The significant differencesofmitochondrial surfacearea beforeandafter ischemia observed inour studyindicatethat myocardial injurywas notfully prevented for up to 83minutesofischemic arrest by the current method of myocardial preservation. Although significant ischemic injury could be found in anyone of the three cardiac chambers, the severity of ischemic injury is widely varied among the chambers. Myocardial preservation is mosteffective in the LV.The

Volume 102 Number 5

Myocardial preservation 68 7

November 1991

percentchange in mitochondrialsizewasonly 14.3%. The RV, with a 43.7% increase in mitochondrial size after ischemic arrest, was less well preserved than the LV. Furthermore, the RA showed an 88% increase in mitochondrialsizeand had the poorestpreservationamong the three chambers. In reviewing the literature we found no previous reports of investigations to compare the effectiveness of myocardialpreservationamong cardiac chambers from the viewpoint of ultrastructural morphometric study.Our present study demonstrates the disproportionate effectiveness of myocardial preservation among the three cardiac chambers during cardioplegic arrest from the viewpoint of an ultrastructural morphometric study. The selective and differential natures of chamber injury and ability to precisely quantitate myocardial injury couldfurther support the previoushypothesis that warmer temperatures in the right side of the heart may induce more severe ischemic injury.I4-16 The potential weaknessof this study may be its lack of temperature measurement to correlate whether the chamber with more severe ischemic injury had a higher myocardialtemperature. Our recent work'? has demonstrated uneven myocardial hypothermia among three cardiac chambers (RA, RV, and LV) during cardioplegic arrest in 55 consecutive patients. The RA had the highest temperature (19.1 0 ± 4.10 C), whereas temperature in the RV (12.7 0 ± 4.8 0 C) was lower than in the RA but significantly higher than in the LV (7.3 0 ± 3.40 C). Becausethe patients in this study wereall providedthe samemethod of myocardial preservation and operated on by the same surgical team, the lack of temperature measurementwas not considereda significantweaknessof this study. Although cardioplegic arrest can offer effective preservationof the LV if the aortic clamp time is limited, our results suggest that the myocardial injury after an average of 83 minutes of ischemicarrest is more severein the right-sided heart chambers than in the LV, with the RA havingthe poorest preservation. Ultimately, because the poorereffectiveness of myocardialpreservationappears in the right-sidedheart chambers, the right sideof the heart should not be neglected in our continuing efforts to improve myocardial preservation. We acknowledge Hwa-Mei Liu for her secretarial assistance in the preparation of the manuscript and Ju-Shin Hau for her technical assistance. .REFER ENCES I. Brazier J, Hottenrott C, Buckberg GO. Non-coronary collateral myocardial blood flow. Ann Thorac Surg 1975; 19:426-35. 2. Rosenfeldt FL, Watson DA II. Interference with local

3.

4.

5.

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

myocardial cooling by heat gain during aortic cross-clamping. Ann Thorac Surg 1979;27:13-6. Novick RJ, Stefaniszyn HJ, Morin JE, et al. Atrial electrical activity and its suppression during cardioplegic arrest in pigs. J THORAC CARDIOVASC SURG 1983;86:235-41. Tchervenkov CI, Wynands JE, Symes JF, et al. Persistent atrial activity during cardioplegic arrest: a possible factor in the etiology of post-operative supraventricular tachyarrhythmias. Ann Thorac Surg 1983;36:437-43. Rabinovitch MA, Elstein J, Chiu RCJ, Rose CP, Arzoumanian A, Burgess JH. Selective right ventricular dysfunction after coronary artery bypass grafting. J THORAC CARD10VASC SURG 1983;86:444-6. Silva R, Chen YF, Sell TL, Lowe JE, Jones RH. Recognition of reversible and irreversible myocardial injury by technetium pyrophosphate extraction kinetics. J THORAC CARDIOVASC SURG 1987;94:104-9. Morris JJ III, Hamm DP, Pellom GL, Abd-Eifattah A, Wechsler AS. Differential ventricular ischemic injury: an experimental model of right ventricular failure with a variable degree ofleft ventricular dysfunction. J THORAC CARD10VASC SURG 1988;96:590-9. Schaper J. Morphological changes in the human heart caused by global ischemia during open heart surgery. In: Roberts AJ, ed. Myocardial protection in cardiac surgery. New York: Marcel Dekker, 1987:53-69. Chen YF, Lin YT. Comparison of blood cardioplegia to electrolyte cardioplegia on the effectiveness of preservation of right atrial myocardium: mitochondrial morphometric study. Ann Thorac Surg 1985;39:134-9. Trump BF, Mergner WJ, Kahng MW, Saladino AJ. Studies on the subcellular pathophysiology of ischemia. Circulation 1976;53(Pt 1):117-26. Kubler W, Speickerman PG. Regulation of glycolysis in the ischemic and anoxic myocardium. J Mol Cell Cardiol1970; 1:351-9. Buckberg GO. Studies of controlled reperfusion after ischemia. I. When is cardiac muscle damaged irreversibly? J THORAC CARDIOVASC SURG 1986;92:483-7. Whalen DA, Hamilton DG, Ganote CE, Jenning RB. Effect of a transient period of ischemia on myocardial cells. I. Effect on cell volume regulation. Am J Pathol 1974; 74:381-98. Christakis GT, Fremes SE, Weisel RD, et al. Right ventricular dysfunction following cold potassium cardioplegia. J THORAC CARDIOVASC SURG 1985;90:243-50. Fisk RL, Ghaswalla 0, Guilbeau EJ. Asymmetrical myocardial hypothermia during hypothermic cardioplegia. Ann Thorac Surg 1982;34:318-23. Gonzalez AC, Brandon TA, Fortune RL, et al. Acute right ventricular failure is caused by inadequate right ventricular hypothermia. J THORAC CARDIOVASC SURG 1985; 89:386-99. Chen YF, Chen JS, Wang JR, Chiu CC, Lin YT. Uneven myocardial hypothermia among cardiac chambers during hypothermic myocardial preservation. Eur J Cardiothorac Surg 1990;4:618-23.