Circ J 2002; 66: 450–452

Mitral Valve Surgery Under Perfused Ventricular With Moderate

Kazuhito Imanaka, MD; Shunei Kyo, MD; Masanori Ogiwara, MD; Hiroaki Tanabe, MD; Hiroshi Ohuchi, MD; Haruhiko Asano, MD; Yuji Yokote, MD; Satoshi Gojo, MD; Masaaki Kato, MD

The safety and myocardial protective effect of perfused ventricular fibrillation (VF) under moderate hypothermia were investigated. Through a midline sternotomy and opening the left atrium from the right side, isolated surgery was performed under aortic cross-clamping (ACC) and cardioplegic arrest using Bretschneider HTK solution in 96 patients, and under perfused VF in 20 patients (VF Group). Patient characteristics, clinical outcomes, and perioperative variables were compared. A satisfactory surgical view was obtained in all VF Group patients. Patient characteristics in the 2 groups were similar, and both groups had comparable results for mortality and morbidity, operation time, cardiopulmonary bypass time, peak levels of creatine kinase (CK) and its myocardial fraction, hours of mechanical ventilation, intensive care unit stay, and postoperative left ventricu- lar ejection fraction. Even in VF Group patients with preoperative critical hemodynamic compromise, inotropes could be discontinued within 3 days. Thus, no detrimental effect of perfused VF was observed. On the other hand, in patients who underwent ACC and cardioplegic arrest of 120min or longer, peak levels of CK and its myocardial fraction were significantly higher than those of the rest of C group patients and VF Group patients. Perfused VF under moderate hypothermia can be a good alternative myocardial protection strategy during mitral valve surgery, particularly in patients in whom ACC is unsuitable or the duration of ACC is expected to be long. (Circ J 2002; 66: 450–452) Key Words: Mitral valve; Myocardial protection; Perfused ventricular fibrillation

urrently, the most common method of myocardial because of the presence of critical hemodynamic compro- protection during mitral valve surgery is induction mise in 4, severely atherosclerotic ascending aorta in 7, C of electromechanical arrest by aortic cross-clamp- functioning previous coronary artery bypass graft in 3, ing (ACC) and infusion of cardioplegic solution. However, dense adhesion in 3, and for miscellaneous reasons in 3. VF this method is unsuitable in some cases, especially in poor- was usually induced just by systemic cooling. risk patients, and we have used an alternative strategy, In C Group patients, 2,000ml of Bretschneider HTK namely, ventricular fibrillation (VF) without ACC. In this solution at 4°C was infused through the aortic root at a rate study, the safety and efficacy of perfused VF under moder- of around 250 ml/min. When the duration of ACC ex- ate hypothermia were investigated. ceeded 90–120min,1–3 700–1,000ml of high-potassium (20 mmol/L) blood cardioplegic solution, comprising CPB perfusate and additional potassium chloride, was infused Methods every 30min. Patient characteristics, clinical outcomes, and From February 1997 until September 2000, 116 patients variables during and after surgery were compared between underwent isolated mitral valve surgery through a midline the 2 groups. sternotomy and opening of the left atrium from the right side. Patients who required concomitant proce- Statistical Analyses dures or coronary artery bypass surgery were excluded. Data are expressed as mean±standard deviation. Student’s Surgery was performed under ACC and cardioplegic arrest t-test or chi-squared test were used to compare the groups. in 96 patients (C Group), or under VF either without ACC Correlation was analyzed with Pearson’s method. Differ- or with a very short duration of ACC in 20 patients (VF ences were regarded as statistically significant at p<0.05. Group). Cardiopulmonary bypass (CPB) flow was main- tained above 2.4L·min–1·m–2 and the rectal temperature was lowered to 28–30°C. Results In VF Group patients, perfused VF was employed Patient characteristics are summarized in Table1. The 2 groups were comparable in many aspects. The majority of (Received November 26, 2001; revised manuscript received January patients had suffered severe cardiac symptoms. Four VF 9, 2002; accepted January 22, 2002) Group patients (acute myocardial and papillary Department of Cardiovascular Surgery, Saitama Medical School, muscle rupture, 1 case; acute massive mitral regurgitation Saitama, Japan Mailing address: Kazuhito Imanaka, MD, Department of Cardiovas- caused by infectious (IE), 3) were in a severely cular Surgery, Saitama Medical School, 38 Morohongo, Moroyama- impaired general and hemodynamic condition (evident machi, Iruma-gun, Saitama 350-0495, Japan. E-mail: imanaka@ lung edema with hypotension (<80mmHg) and oliguria). saitama-med.ac.jp Mitral valve repair was undertaken in 41% of C Group

Circulation Journal Vol.66, May 2002 Mitral Valve Surgery Under VF 451

Table 1 Patient Characteristics

VF Group (n=20) C Group (n=96) p value MS:MR 4:16 27:69 NS M:F 12:8 43:53 NS Age (years) 55.4±15.2 57.4±12.5 NS BSA (m2) 1.55±0.25 1.51±0.20 NS NYHA II:III:IV 2:9:9 12:56:28 NS Sternal re-entry 7 19 NS MICS 2 18 NS IE 4 8 NS Ejection fraction (%) 68.8±12.9 71.0±12.1 NS Left atrial diameter (mm) 49.8±10.6 52.7±11.3 NS MVP:MVR 9:11 38:58 NS Associated procedure TAP 1 17 NS Miscellaneous 2 7 NS MS, mitral stenosis; MR, mitral regurgitation; BSA, body surface area; NYHA, New York Association functional class; MICS, minimally invasive cardiac surgery; IE, ; MVP, mitral valvuloplasty; MVR, mitral valve replacement; TAP, tricuspid annuloplasty.

Table 2 Clinical Results

VF Group (n=20) C Group (n=96) p value Operation time 267±49 259±65 NS CPB time 144±22 129±40 NS ACC time 1.2±3.6 82.6±29.0 <0.05 Rectal temperature 28.4±1.3 29.5±2.9 NS Peak CK (IU/L) 1,008±640 1,238±977 NS Peak CKMB (IU/L) 94±57 87±46 NS Mechanical ventilation time (h) 11.0±12.7 10.7±11.6 NS Intensive care unit stay (days) 2.0±1.3 2.4±1.1 NS Postoperative ejection fraction (%) 62.7±18.8 62.5±13.2 NS Hospital 2 3 NS Low output syndrome 0 1 NS Cerebral infarction 1 3 NS CPB, cardiopulmonary bypass; ACC, aortic cross clamp; CK, creatine kinase; CKMB, myocardial fraction of CK. patients and 45% of VF Group patients. required large doses of inotropes just after surgery. Peri- Clinical results are shown in Table2. Satisfactory expo- operative stroke occurred in 1 VF Group patient and 3 C sure of the mitral apparatus was obtained in all VF Group Group patients. The 2 groups had a comparable duration of patients by somewhat extending the atriotomy. During mechanical ventilation and intensive care unit stay, and treatment in the vicinity of the antero-lateral commissure, the left ventricular ejection fraction (LVEF) at hospital traction caused aortic regurgitation and ACC was temporar- discharge was almost the same. The New York Heart Asso- ily necessary in 3 patients. Operation and CPB times were ciation functional class of all survivors was I or II. almost the same in the 2 groups. ACC time was signifi- cantly shorter in the VF Group. Peak levels of serum creatine kinase (CK) and its myo- Discussion cardial fraction (CKMB) in the 2 groups were almost the Intraoperative myocardial protection is a major concern. same. The VF time of VF Group patients was 104±26min Induction of cardioplegic arrest with ACC is a widely (range, 52–146). VF time did not show a significant corre- accepted solution for lessening myocardial damage, but it lation with peak CK or CKMB. On the other hand, peak has several potential drawbacks; namely, time limitation, CK was 1,858±1,651 and CKMB was 139±49 in 9 patients reperfusion injury, myocardial edema and diastolic dys- with ACC and cardioplegic arrest longer than 120 min, function.4–7 Moreover, ACC is undesirable in the presence both significantly higher than those of the rest of C Group of atherosclerotic aorta,8 previous coronary bypass grafts9–11 patients and VF Group patients. and so forth. Two VF Group patients died, one of stroke and the other Mitral valve surgery under VF and intermittent ACC has of uncontrollable IE. Three occurred in the C Group: been recognized a suboptimal strategy and was almost low output syndrome in 1 case, rupture of an abdominal abandoned after the undesirable physiology of perfused VF mycotic aneurysm in 1, and uncontrollable IE in 1. Hospi- and poor postoperative hemodynamic results were re- tal mortality of any cause and of cardiac cause was, re- ported,12–14 although these drawbacks could be reduced.15,16 spectively, 10% and 5% in VF Group and 3.1% and 2.1% In the present study, the clinical results of both groups in C Group. In the VF Group, severe postoperative low were comparable, although perfused VF was used in output syndrome was not observed and prolonged (4 days patients whose clinical condition was more complicated. or longer) infusion of inotropic agent was unnecessary, The levels of CK and CKMB were almost the same, which although patients with preoperative suggested that perfused VF was not an inappropriate myo-

Circulation Journal Vol.66, May 2002 452 IMANAKA K et al. cardial protection strategy. Thus, we believe that VF under ACC or none at all was used in a selected group of patients. moderate hypothermia without ACC or with ACC of short The clinical results were comparable with those of the duration is a safe and good alternative during mitral valve control group patients, which proved the safety and satis- surgery. factory myocardial protective effect of this strategy. When Although this classical strategy may appear to be anach- cardioplegic arrest with ACC is inappropriate or the dura- ronistic, it has several advantages. Complex valvuloplasty tion of ACC is expected to be long, perfused VF under or repair of annular abscesses sometimes requires consider- moderate hypothermia can be a good alternative. able time but in cases of cardioplegic arrest, time is limited and overzealous infusion of cardioplegic solution can be References harmful.17 If perfused VF is adopted in such cases, however, as much time as necessary can be used for treatment of the 1. Bretschneider HJ, Hubner G, Knoll D, Lohr B, Nordbeck H, Spieckermann PG. Myocardial resistance and tolerance to : lesions. In the present study, mean VF time of the VF Physiological and biochemical basis. J Cardiovasc Surg 1975; 16: Group patients exceeded 100min and the longest VF time 241–260. was 146min, but such extended period of VF did not cause 2. Stapenhorst K. Prolonged safe ischemic using hypo- severe myocardial damage when combined with adequate thermic Bretschneider cardioplegia combined with topical cardiac cooling: Clinical experience with cardiac arrest of up to 180min in a perfusion pressure at moderate hypothermia. few borderline cases. Thorac Cardiovasc Surg 1981; 29: 272–274. In the VF Group patient with perioperative stroke, 3. Gebhard MM, Bretschneider HJ, Gersing E, Preusse CJ, Schnabel surgery was carried out without ACC. The merits of VF PA, Ulbricht LJ. Calcium-free cardioplegia-Pro. Eur Heart J 1983; in preventing cerebral emboli were not demonstrated in 4(SupplH): 151–160. 4. Foglia RP, Steed DL, Follette DM. Iatrogenic myocardial edema with the present study, because stroke is not always caused by potassium cardioplegia. J Thorac Cardiovasc Surg 1979; 78: 217– plaque in the aorta, and because we did not completely 222. avoid ACC even in VF Group patients. Theoretically, how- 5. Hsu DT, Weng ZC, Nicolosi AC, Detwiler PW, Sciacca R, Spotnitz ever, these merits are obvious, and omitting ACC can be HM. Quantitative effects of myocardial edema on the left ventricular quite advantageous in some settings.9 pressure-volume relation. J Thorac Cardiovasc Surg 1993; 106: 651–657. Another advantage that is difficult to demonstrate quan- 6. Amirhamzeh MMR, Dean DA, Jia CX, Cabreriza SE, Starr JP, Sardo titatively is the coaptation test during mitral valvuloplasty. MJ, et al. Iatrogenic myocardial edema: Increased diastolic compli- The whole cardiac structure under ACC and potassium ance and time course of resolution in vivo. Ann Thorac Surg 1996; arrest is completely flaccid and the coaptation test is liable 62: 737–743. 7. Casthely PA, Shah C, Mekhjian H, Swistel D, Yoganathan T, Komer to be unreliable, especially in cases in which the left ven- C, et al. Left ventricular diastolic function after coronary artery tricle is very large. Moreover, we occasionally encounter bypass grafting: A correlative study with three different myocardial unexpected ST-T changes, probably resulting from air protection techniques. J Thorac Cardiovasc Surg 1997; 114: 254– emboli in the caused by rapid saline injec- 260. 8. Blauth CI, Cosgrove DM, Webb BW, Ratliff NB, Boylan M, tion into the left . Under perfused VF, the integrity Piedmonte MR, et al. Atheroembolism from the ascending aorta: An of the annular structure is preserved, and coaptation can be emerging problem in cardiac surgery. J Thorac Cardiovasc Surg checked just by pushing the aortic root gently and produc- 1992; 103: 1104–1112. ing aortic regurgitation. We believe that the coaptation test 9. Byrne JG, Aranki SF, Adams DH, Rizzo RJ, Couper GS, Cohn LH. under perfused VF is more reliable, with a much lower risk Mitral valve surgery after previous CABG with functioning IMA grafts. Ann Thorac Surg 1999; 68: 2243–2247. of air embolism. 10. Grondin CM, Pomar JL, Hebert Y, Bosch X, Santos JM, Enjalbert On the other hand, some aspects require caution when M, et al. Reoperation in patients with patent atherosclerotic coronary perfused VF is used. Perfusion pressure should be managed vein grafts: A different approach to a different . J Thorac with care because low perfusion pressure is a major hin- Cardiovasc Surg 1984; 87: 379–385. 12 11. Marshall WG Jr, Saffitz J, Kouchoukos NT. Management during drance to this strategy. In patients in whom more than a reoperation of aortocoronary saphenous vein grafts with minimal mild degree of aortic regurgitation coexists with mitral valve atherosclerosis by angiography. Ann Thorac Surg 1986; 42: 136–137. disease, perfused VF is not applicable because backflow 12. Hottenrott C, Maloney JV Jr, Buckberg GD. Studies of the effects of disturbs the surgical field and hampers proper treatment. If ventricular fibrillation on the adequacy of regional myocardial flow. III. Mechanisms of ischemia. J Thorac Cardiovasc Surg 1974; 68: the aortic valve is left untouched and ACC is undesirable in 634–645. such cases, deep hypothermia and low flow perfusion9 is 13. Grover FL, Fewel JG, Ghiogoni JJ, Norton JB, Arom KV, Trinkle recommended. JK. Effects of ventricular fibrillation on coronary blood flow and myocardial metabolism. J Thorac Cardiovasc Surg 1977; 73: 616– 624. Study Limitations 14. Maloney JV Jr, Cooper N, Mulder DG, Buckberg GD. Depressed This is a retrospective observational study. Patients were cardiac performance after mitral valve replacement: A problem of not randomized for intraoperative myocardial management. myocardial preservation during operation. Circulation 1975; 52: I- In addition, the number of the VF Group patients was 3–I-8. small. Therefore, the superiority of either strategy can not 15. Grotte GJ, Levine FH, Kay HR, Fallon JT, Austen G, Buckley MJ. Effect of ventricular fibrillation and potassium-induced arrest on be documented from the present study and a prospective myocardial recovery in hypothermic heart. Surg Forum 1980; 31: randomized study of a large patient group is warranted. 296–298. 16. Vinas JF, Fewel JG, Arom KV, Trinkle JK, Grover FJ. Effects of systemic hypothermia on myocardial metabolism and coronary blood Conclusions flow in the fibrillating heart. J Thorac Cardiovasc Surg 1979; 77: 900–907. As a strategy for myocardial protection during mitral 17. Kofsky ER, Julia PL, Buckberg GD. Overdose reperfusion of blood valve surgery, perfused VF with either a short period of cardioplegia. J Thorac Cardiovasc Surg 1991; 101: 275–283.

Circulation Journal Vol.66, May 2002