ORIGINAL ARTICLE

     SPINAL PROTECTION WITH VENO-ARTERIAL BY-PASS FOR SURGERY OF
      THORACIC AORTA:26-YEAR EXPERIENCE

    THE IMPACT OF GRAFT SELECTION ON OUTCOME OF CORONARY
      BYPASS SURGERY AND INFLUENCE OF SURGICAL EXPERIENCE

    THE IMPACT OF GRAFT SELECTION ON OUTCOME OF CORONARY
      BYPASS SURGERY AND INFLUENCE OF SURGICAL EXPERIENCE

Yukihiro Kaneko, Bernard Schlechta, Werner Steinberger, Irene Agstner*, Ernst Wolner, Werner Mohl. Departments of Cardiothoracic Surgery, and Surgical Documentation and Biometrics*, University of Vienna, Austria

   ABSTRACT

   Introduction

   Materials and Methods

  Table 1. Variables of patients’ characteristics and surgical technique.
      Frequencies are shown by percentages
  Table 2. Patient characteristics identified as incremental risk factors.
  Results

   DISCUSSION
   References

ABSTRACT

In a retrospective study of 787 consecutive patients after first-time isolated coronary bypass grafting by a single surgeon up to 16 years, the risk ratios of arterial grafting and sequential saphenous vein (SV) grafting on overall mortality, mortality related to cardiac disease, and adverse cardiac events were quantified by univariate and multivariate analyses corrected for the influence of preoperative characteristics. Arterial grafting was an incremental risk factor in the first 5 years, but decremental risk factor in the later 11 years. The risk ratio of sequential SV graft to the left coronary artery on overall mortality (1.30, 95 % CI, 0.93-1.81) was significantly higher than that to the right coronary artery (0.64, 95 % CI, 0.42-0.99). The outcome of arterial grafting significantly improved over time, but outcome of sequential SV grafting to the right coronary artery did not. Surgical volume did not influence the outcome. (Heart Views. (2002;3(3): 118-123) © 2002 Gulf Heart Association

(Heart Views. 2002;3(3):118-123) © 2002 Gulf Heart Association.

Key Words:

   coronary disease   surgery    revascularization

    survival   risk factors.

Introduction

In coronary bypass surgery, several surgical factors including selection of graft material, selection between single grafting and sequential grafting, myocardial protection technique, suture materials, suturing technique, and surgeon’s dexterity may influence the clinical outcome. After stratifying for gender, preoperative New York Heart Association (NYHA) functional class, age, preoperative myocardial infarction, and extent of coronary stenosis, this study examines the influence of: 1) combined arterial/venous grafting versus exclusive venous grafting, 2) use of sequential venous graft versus single venous graft, 3) surgeon experience, 4) surgical volume, and 5) evolution of cardioplegic strategy in patients undergoing first time isolated coronary bypass grafting.

Materials and Methods

We made a retrospective study on long-term outcome of 787 consecutive patients after first-time isolated coronary bypass grafting between 1984 and 1999 by a single surgeon (W.M.). Five preoperative characteristics were inputted in the dataset by reviewing the hospital records. The preoperative characteristics included: gender, preoperative NYHA functional class, age at operation, history of preoperative myocardial infarction (MI), and extent of coronary stenosis (single, double, and triple vessel diseases without left main stenosis were respectively scored as one, two and three; and left main stenosis was scored as four). Use of left internal thoracic artery (ITA) and/or other arterial grafts, and sequential use of saphenous vein (SV) graft was inputted in the dataset by reviewing operation record. Y graft was considered as a variant of sequential graft. Sequential SV grafts were classified according to the coronary artery to which they were anastomosed. When a graft was anastomosed to the left coronary artery and/or its tributaries, it was classified as left-sided. When a graft was anastomosed to the right coronary artery and/or its tributaries, it was classified as right-sided. When a graft was anastomosed to both the right and left coronary systems, it was classified as both left-sided and right-sided. The follow-up data of the patients were collected from hospital record and by telephone questionnaire to the patient or a family member. At surgery, cardiopulmonary bypass with normal or mildly decreased body temperature was used. Antegrade crystalloid cardioplegia was used for myocardial protection until 1991. Combined antegrade/retrograde blood cardioplegia was used thereafter (1,2). Use of left ITA graft, introduced in 1986 in our institution, became a routine from 1989 (3). The types of grafts, ITA or SV, as well as single or sequential, were at the discretion of the surgeon. Only the greater saphenous vein was used as vein grafts. The SV taken from the lower leg was preferred to that taken from the upper leg. Use of a circular SV graft was discouraged (4). Postoperatively, oral aspirin of 100 mg was routinely administered and continued indefinitely. Statistical analysis was made by the proportional hazards model of Cox to model and to quantify the impact of grafting techniques corrected for the influence of five preoperative characteristics using SAS-software (SAS Institute Inc., Cary, NC) (5). Univariate and multivariate analyses were performed on three endpoints: overall mortality including hospital death, mortality clearly related to heart disease, namely cardiac mortality, and adverse events (i.e. death, MI or re-intervention). Multivariate analysis was performed by two different modes: one assuming that right-sided and left-sided sequential SV graft to be identical, distinguishing two grafting techniques (arterial graft, and sequential SV graft to either coronary artery) from single SV graft; and the other assuming right-sided and left-sided sequential SV graft to be different grafting techniques, distinguishing three grafting techniques (use of arterial graft, left-sided sequential SV graft, and right-sided sequential SV graft) from single SV graft. It is plausible that surgical outcome may be influenced by surgical experience, surgical volume and evolution of cardioplegia technique (6-9). Surgical experience, especially on the use of arterial graft, might be limited until 1988, and thereby might influence the outcome. To explore the influence of surgical experience, the cohort was divided into two groups: the patients operated on until 1988 when ITA use was not routine (n = 263), and the patients operated on after 1988 when the left ITA was routinely used unless contraindicated (n = 524). Multivariate analysis assuming the right-sided and left-sided sequential SV graft to be different grafting techniques was performed in each group, and risk ratios of grafting techniques were calculated against single SV grafts in each group. Yearly number of coronary bypass surgery was 51 or less, i.e., less than one case per week, in 1984, ’92, ’93, ’95, ’97 and ’98, whereas it was 52 or more in the remaining years. To detect the influence of surgical volume, the risk ratios of surgical factors of patients operated on in six low surgical-volume years above-mentioned (n = 209; mean yearly volume 34.8 patients/year) were compared with those in the remaining ten high surgical-volume years (n = 578; mean yearly volume 57.8 patients/year). To detect the influence of cardioplegia technique, comparison was made between surgical years until 1991 when antegrade crystalloid cardioplegia was used (n = 452), and surgical years after 1991 when combined antegrade/retrograde blood cardioplegia was used (n = 335). All p values were given two-tailed. P value of less than 0.05 was considered statistically significant.

Results

Patient characteristics and surgical factors are shown in Table 1. Arterial graft/grafts were used in 526 of 787 patients; the left ITA in all 526 patients, right ITA in 26 patient, gastroepiploic artery in 2 patients, and radial artery in 1 patient. Sequential SV graft/grafts were used in 304 patients. The overall mortality was 3.7 % at 1 month, 9.0 % at 1 year, and 19.9 % at 5 years. The cardiac mortality was 3.2 % at 1 month, 7.3 % at 1 year, and 11.9 % at 5 years. The rate of adverse events was 3.7 % at 1 month, 12.9 % at 1 year, and 25.4 % at 5 years. Univariate and multivariate analyses in the total cohort identified several preoperative characteristics as incremental risk factors of overall mortality, cardiac mortality, and adverse events. (Table 2) Oddly, use of arterial graft was identified as incremental risk factor, whereas right-sided sequential graft as decremental risk factor. Risk ratios of left-sided and right-sided SV graft on overall mortality were statistically different. (Fig. 1) The influences of surgical experience on different grafting techniques are depicted in Figure 2. The use of arterial graft was incremental risk factor on overall mortality, cardiac mortality, and adverse events in patients operated on until 1988,

Table 1. Variables of patients’ characteristics and surgical technique. Frequencies are shown by percentages

Table 2. Patient characteristics identified as incremental risk factors.

whereas it became decremental risk factor on cardiac mortality and adverse events after 1988. Statistically significant improvement in the risk ratio of the arterial graft was observed in all three endpoints. Risk ratio of left-sided sequential SV graft on overall mortality decreased significantly. Risk ratios of left-sided sequential SV graft on cardiac mortality and adverse event tended to decrease (statistically not significant). There appeared no change in the risk ratio of right-sided sequential SV graft on any endpoints. Influence of surgical volume on risk ratios of different grafting techniques are shown in Figure 3. Surgical volume did not appear to influence the risk ratios of any grafting technique. Comparison of risk ratios between the years when antegrade crystalloid cardioplegia was used and the years when combined antegrade/retrograde blood cardioplegia was used identified no statistically significant change. However, slight improvement was seen in risk ratios of all three grafting techniques after revising myocardial protection technique (Figure. 4).

DISCUSSION

The previous literature documented that female gender, older age, extensive coronary stenosis and history of MI are predictors of unfavorable long-term outcome (10,11). The result of the present study shows preoperative risk factors in agreement with the literature. It has been established from numerous studies that the use of left ITA graft improves the long-term outcome (10,11). In the present study, however, the use of arterial graft was an incremental risk factor despite the fact that left ITA graft was used in all the patients receiving arterial graft. The outcome of arterial grafting became favorable after 1988, but the unfavorable result of arterial graft until 1988 has not been offset yet by the favorable result after 1988. The discrepancy between the literature and the present study might have been induced by a combination of three factors i.e. myocardial protection technique that attenuates reperfusion injury, patient selection bias, and the learning curve in arterial grafting (1). Until 1988, use of arterial graft was not familiar to the institution and the surgeon. This might have negatively influenced the outcome of arterial grafting in the period until 1988. It should be noted that even in the period until 1988 when the use of arterial graft showed unfavorable results in long-term follow up, use of arterial graft improved the early outcome up to three months. (Figure 5). The longer the follow-up period, the worse the impact of arterial grafting became. The impact became constant after the follow-up period of 12 months.
other hand, the positive impact of arterial grafting became stronger as the follow-up period elongates, and it became constant after 18 months of follow-up. Younger surgeons receives increasing amount of education about arterial grafting. It is anticipated that their personal risk ratio of arterial grafting is better than the result in this study. Nevertheless, the use of arterial graft does not necessarily improve the personal result. We recommend, therefore, that young surgeons should monitor the personal surgical result of ITA grafting for at least 18 months so that each surgeon can be aware of the personal risk ratio of ITA grafting.

FIG-1

Fig. 1. Risk ratios of different types of graft in multivariate analysis are shown by a cross. Risk ratios of arterial graft were calculated assuming the right-sided and left-sided sequential grafts to be different. The vertical bar indicates actual value, and the transverse bar and numerals indicate 95 % confidence limit. Asterisks indicate statistical significance. SSVG = sequential saphenous vein graft.

FIG-2

Fig. 2. Influence of surgical experience on risk ratios of different types of graft. Comparison was made between the first 5 years when arterial grafting was not routinely used and the remaining 11 years when use of the left ITA became routine. Asterisks indicate statistical significance between the different periods.

A review of the literature revealed diverse results as to the impact of sequential SV grafting on graft patency and clinical outcome. Some indicated better patency and favorable clinical outcome in patients with sequential SV grafts (12-19), and others reported vise versa (20-22). The present study raised two suggestions that may account for the diversity of the previously reported results of sequential SV grafting. First, the impact of right-sided sequential SV grafts on long-term outcome is probably different from that of left-sided ones. Second, surgical experience may have different influences on left-sided and right-sided sequential SV grafts. The diversity of the impacts of sequential SV grafts in the literature may be explained by different distribution of the coronary artery to which sequential grafts were anastomosed and/or by different surgical experience of the surgeons. The reason why surgical experience improved the risk ratios of arterial grafting significantly while it did not influence the risk ratios of right-sided sequential SV grafting is unknown. It may be simply because we set the chronological threshold of surgical experience based on routine use of arterial graft. Perhaps, the more the grafting technique is difficult, the greater the influence of surgical experience. Surgical volume did not influence the impact of the use of arterial graft or use of sequential SV graft. However, we should cautiously interpret the result since there may be a chance of false-negative result if the cohort of this study is too small, or if the threshold set to divide the cohort is inappropriate. Evolution of cardioplegic technique appeared to have minimal but consistent favorable influence on the impacts of arterial graft and sequential SV graft. The result may be modified by the influence of surgical experience as the period when surgical experience was considered insufficient is covered by the period when antegrade crystalloid cardioplegia was used. With the chronological threshold at the year 1991 when the cardioplegia technique was changed, similar improvement in the risk ratios was observed of all the three types of grafting. Therefore, we assume that this improvement represents the impact of cardioplegia technique evolution, not of accumulating surgical experience that is likely to be grafting technique specific.
Several reports indicated that the later year of surgery was a favorable predictor of long-term outcome (11,21). It has not been clear whether this finding is due to learning curve of individual surgeon, technical/technological progress, or better surgical education given to subsequent generation of surgeons. The present study indicated that learning effect and evolution of cardioplegia technique might contribute towards improvement in the long-term outcome after coronary surgery.

FIG- 3

Fig. 3. Influence of surgical volume on risk ratios of different types of graft. Yearly number of coronary bypass cases was below 52 in 1984, ‘92, ‘93, ‘95, ‘97, and ‘98 (low volume), whereas it was over 53 in the remaining years (high volume). aa

FIG- 4

Fig. 4. Comparison of risk ratios of different types of graft between the period when antegrade crystalloid cardioplegia was used (1984-1991) and the period when combined antegrade/retrograde blood cardioplegia was used (1992-1999).

The present study raised a question as to the modern coronary surgery that recently adopted off-pump coronary surgery. When performing off-pump surgery, whole patient management including anesthesia, myocardial protection, graft material selection, proximal anastomosis site of free graft, target coronary location, and attitude toward incomplete revascularization have to adapt for the new surgical technique. Consequently, the positive impact of arterial grafting confirmed in the on-pump coronary bypass may not be guaranteed in the off-pump coronary bypass particularly in its initial phase. Although in-situ arterial grafts are more readily handled than free vein grafts in off-pump surgery, unconditioned assumption that ITA grafts are decidedly favorable should be avoided. There are several limitations inherent to this study. The decisions whether to use arterial grafts and sequential SV grafts were made arbitrarily by the surgeon. The decision may have been influenced by coronary disease severity and coronary artery anatomy including disposition of target vessel and dominancy. Coronary disease severity and coronary anatomy may have influenced the outcome, and thereby may have introduced bias into the result. In addition, the study is based on experience of a left-handed surgeon who performs surgery from the patient’s left. The surgical exposure is different when the surgeon stands on the patient’s left and when the surgeon stands on the patient’s right. This possibly influenced technical difficulty and thereby outcome. Surgical technique, myocardial protection technique, anesthetic technique, and postoperative management changed over time during the study period. These changes probably influenced the factors including likelihood of graft and coronary artery spasm, fluctuation in the blood pressure and heart rate, exposure of the target vessels. Consequently, the result of this study may not be reproducible with different surgeon’s handedness, surgical technique, myocardial protection technique, anesthetic technique, and postoperative management. The well-known preoperative risk factors including left ventricular ejection fraction, history of hypertension, history of smoking, and serum cholesterol were not inputted in the dataset, allowing it to be less optimal. Moreover, coronary angiography was only indicated when clinically necessary because of financial reasons. Consequently, lack of information about graft patency made the interpretation of the result to be less accurate allowing only insight into clinical parameters such as morbidity and mortality as well as quality of life.

References

1. Simon P, Mohl W, Neumann F, Owen A, Punzengruber C, Wolner E. Effect of coronary artery bypass grafting on global and regional myocardial function. An i ntraoperative echocardiographic assessment. J Thorac Cardiovasc Surg 1992; 104: 40-5.

2. Newmann F, Mohl W, Griesmacher A, Simon P, Zweytick B, Kupilik N, Stix G, Moidl R, Wolner E. Perioperative myocardial injury with different modes of antegrade and retrograde cardioplegic delivery. Eur J Cardiothorac Surg 1996; 10: 185-93.

3. Izzat MB, West RR, Bryan AJ, Angelini GD. Coronary artery bypass surgery: current practice in the United Kingdom. Br Heart J 1994; 71: 382-5.

4. Grondin CM, VouchÈ P, Bourassa MG, LespÈrance J, Bouvier M, Campeau L. Optimal patency rates obtained in coronary artery grafting with circular vein grafts J Thorac Cardiovasc Surg 1978; 75: 161-7.

5. Cox DR. Regression models and life tables. J R Stat Soc 1972; 34: 187-202.

6. Sollano JA, Gelijns AC, Moskowitz AJ, Heitjan DF, Cullinane S, Saha T, Chen JM, Roohan PJ, Reemtsma K, Shields EP. Volume-outcome relationships in cardiovascular operations: New York State, 1990-1995. J Thorac Cardiovasc Surg 1999; 117: 419-28.

7. Hannan EL, Kilburn H Jr, Bernard H, O’Donnel JF, Lukacik G, Shields EP. Coronary artery bypass surgery: the relationship between inhospital mortality rat eand surgical volume after controlling for clinical risk factors. Med Care 1991; 29: 1094-107.

8. Jacquet LM, Noirhomme PH, Van Dyck MJ, El Khoury GA, Matta AJ, Goenen MJ, Dion RA. Randomized trial of intermittent antegrade warm cardioplegia versus cold crystalloid cardioplegia. Ann Thorac Surg 1999; 67: 471- 7.

9. Ronne T, Pehkonen E, Kaukinen S, Tarkka M. Comparison of cardioprotection with crystalloid and blood cardioplegia in CABG patients. J Cardiothorac Vasc Anesth 1993; 7: 679-83.

10. Eagle KA, Guyton RA, Davidoff R, et al. ACC/AHA guidelines for coronary artery bypass graft surgery: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee to Revise the 1991 Guidelines for Coronary Artery Bypass Graft Surgery). J Am Coll Cardiol 1999; 34: 1262-346.

11. Kirklin JW, Barratt-Boyes BG. Cardiac Surgery: morphology, diagnostic criteria, natural history, techniques, results, and indications. 2nd ed. New York: Churchill Livingstone, 1993.

12. Bigelow JC, Bartley TD, Page US, Krause AH Jr. Long- term follow-up of sequential aortocoronary venous grafts. Ann Thorac Surg 1976; 22: 507-14.

13. Sewell WH, Sewell KV. Technique for the coronary snake graft operation. Ann Thorac Surg; 22: 58-65.

14. Pfluger N, Turina M, Goebel N, Krayenbuhl C, krayenbuhl HP, Kugelmeier J, Moccetti T, Senning A, Steinbrunn W, Rothlin M. Single and sequential aorto-coronary bypass: comparison of patency rate. Schweiz Med Wochenschr 1980; 110: 1649-50.

15. Meurala H, Valle M, Hekali P, Somer K, Frick MH, Harjola PT. Patency of sequential versus single vein grafts in coronary bypass surgery. Thorac Cardiovasc Surg 1982; 30: 147-51.

16. Limet R, david JL, Magotteaux P, Larock MP, Rigo P. Prevention of aorto-coronary bypass graft occlusion. Beneficial effect of ticropidine on early and late patency rates of venous coronary bypass grafts: a double-blind study. J Thorac cardiovasc Surg 1987; 94: 773-83.

17. Paz MA, Lupon J, Bosch X, Pomer JL, Sanz G, the GESIC Study Group. Predictor of early saphenous vein aortocoronary bypass graft occlusion. Ann Thorac Surg 1993; 56: 1101-6.

18. Alderman EL, Levy JH, Rich JB, Nili M, Vidne B, Schaff H. Analysis of coronary graft patency after aprotinin use: results from the international multicenter aprotinin graft patency experience (IMAGE) trial. J Thorac Cardiovasc Surg 1998; 116: 716-30.

19. Christenson JT, Simonet F, Schumuziger M. Sequential vein bypass grafting: tactics and long-term results. Cardiovasc Surg 1998: 6; 389-97.

20. Goldman S, Zadina K, Krasnicka B, Moritz T, Sethi G, Copeland J, Ovitt T, Henderson W. Predictor of graft patency 3 years after coronary artery graft surgery. J Am Coll Cardiol 1997; 29: 1563-8.

21. Sergeant P, Blackstone E, Meyns B, K.U. Leuven Coronary Surgery Program. Eur J Cardiothorac Surg 1997; 12: 1-19. 22. Voors AA, van Brussel BL, Kelder JC, Plokker HW. Usefulness of hyperglyceridemia in predicting myocardial infarction late after coronary artery bypass operation. Am J Cardiol 1997; 79: 1350-4.

Address for correspondence: Yukihiro Kaneko, MD. Dept. of Cardiovascular Surgery, Japanese Red Cross Medical Center, 4-1-22 Hiroo, Shibuya-ku, Tokyo 150-8935, Japan. Tel: +81-3-3400-1311, Fax: +81-3-3409-1604. e-mail: yukihirokaneko@hotmail.com