Ventricular Function and the Native Coronary Circulation Five Years after Myocardial Revascularhation

Transcription

1 Ventricular Function and the Native Coronary Circulation Five Years after Myocardial Revascularhation Denis H. Tyras, M.D., Naseer Ahmad, M.D., George C. Kaiser, M.D., Hendrick B. Barner, M.D., John E. Codd, M.D., and Vallee L. Willman, M.D. ABSTRACT Of 531 patients who underwent coronary artery bypass grafting during 1970 to 1973,181 were restudied by ventriculography and by graft and coronary angiography at least 5 years following operation. Five patterns of postoperative ventricular function were identified: improved ventricular function resulting in normal left ventricular (LV) function; normal ventricular function that was unchanged; abnormal ventricular function that improved but did not reach normal; abnormal ventricular function that remained unchanged; and deterioration of LV function. Patients who regained (40) and those who retained normal ventricular function (49) comprise 49% of the series and patients with deterioration of ventricular function, only 20%. Graft patency and angina relief were significantly better in those with normal LV function than in those with LV deterioration. Progression of disease in grafted coronary arteries was similar in all groups, but was significantly higher in ungrafted coronary arteries (61.3%) in the patients showing deterioration than in either the improved patients or those with an unchanged normal LV (33.3% each) (p < 0.05). The high incidence of progression of disease in ungrafted coronary arteries in the group with deterioration suggests that low graft patency and deterioration of ventricular function in this group might both be related to intrinsic acceleration of coronary atherosclerosis unrelated to operative intervention. Symptomatic relief of angina pectoris following coronary artery bypass grafting (CABG) has From the Departments of Surgery and Medicine, Saint Louis University School of Medicine, 1325 S Grand Blvd, St. Louis, MO. Presented at the Twenty-fifth Annual Meeting of the Southern Thoracic Surgical Association, Nov 2-4, 1978, Marco Island, FL. Address reprint requests to Dr. Tyras, Department of Surgery, Saint Louis University School of Medicine, 1325 S Grand Blvd, St. Louis, MO been reproducibly demonstrated by several investigators [6, 13, 211, but objective, measurable benefits of CABG are less well documented. Recent reports have suggested increased longevity following CABG and have identified standards of operative mortality and graft patency rates which can be achieved at most surgical centers [13, 15,221. In addition, data are now available showing low rates of perioperative [3] and late myocardial infarction in CABG patients [23]. The long-term effects of CABG on left ventricular (LV) function and on the native coronary circulation are among the factors requiring further evaluation. The purpose of this report is to examine LV function and changes in the native coronary circulation in a large group of patients undergoing follow-up studies at least 5 years following CABG. Materials and Methods Isolated CABG was performed in 531 patients at Saint Louis University Hospital between 1970 and 1973 (excluding patients undergoing concomitant valve replacement or ventricular aneurysmectomy or both). Follow-up cardiac catheterization was performed at least 5 years following CABG in 181 patients. An additional 197 patients (Table 1) refused follow-up angiographic study because of either physician or personal preference. Other reasons for lack of follow-up are enumerated in Table 1. Left ventriculography, coronary arteriography, and bypass graft visualization were performed by the Judkins technique [12]. LV function was assessed by evaluating segmental wall motion of five specified ventricular segments viewed in the right anterior oblique projection of the left ventriculogram and by evaluating the overall ejection fraction [9]. Multiple views of the native coronary arteries by Denis H. Tyras

2 548 The Annals of Thoracic Surgery Vol 27 No 6 June 1979 Table 1. Status of 531 Patients Undergoing Coronary Artery Bypass Operation, 1970 to 1973 Status No. of Patients (%) Unavailable Lost to follow-up 27 (5.1) Operative death 21 (3.9) Late death (before 5-year 53 (9.9) catheterization) No 5-year follow-up angiogram Technical and miscellaneous dif- 21 (3.9) ficulties Refusal of study by patients or 197 (37.1) physicians All grafts occluded on previous 31 (5.8) study Follow-up angiogram at 5 or more 181 (34.1) years and bypass grafts were obtained and evaluated in all patients. Ventricular function and the native coronary arteries at 5 years or more after operation were compared with findings of the study performed before operation. Progression of atherosclerotic narrowing in the native coronary circulation was assessed by two observers comparing multiple views of the coronary arteries in the preoperative and 5-year postoperative studies. Any increase in luminal narrowing of 10 /~ or more or development of a new stenosis of 50% or more was considered significant progression. If all major coronary arteries with a stenosis of 50% or more were grafted, the patient was judged to be completely revascularized at operation. Risk factors and perioperative or late myocardial infarction were assessed by criteria previously described [22]. Statistical analysis was performed with chi-square testing. Results Ventricular Function Five patterns of ventricular function were identified among the 181 patients studied at least five years following CABG: improvement resulting in normal LV function, normal ventricular function that was unchanged, abnormal ventricular function that improved but did not reach normal, abnormal ventricular function that remained unchanged, and deterioration of LV function. Among these five groups, the only significant differences in risk factors are found in the incidence of previous myocardial infarction (MI) and in sex distribution (Table 2). There was a significantly (p < 0.05) higher incidence of previous MI in those patients with abnormal postoperative LV function, whether unchanged, improved, or worse than in those with normal postoperative ventricular function. Among the 20 women evaluated, 16 had normal LV function preoperatively, which limited the available postoperative category to normal ventricular function that was unchanged or deterioration of LV function. Angina Relief versus Ventricular Function Continued relief of angina pectoris was high in all groups but was best in the groups that had improvement resulting in normal LV function or had normal ventricular function that was unchanged (Table 3). In these two groups, 78% and 72% of patients, respectively, had complete relief of angina (Canadian Heart Association Angina Class 0 [71). Complete relief of angina pectoris in these two groups was significantly better than in the other three categories of ventricular function (p < 0.001). The degree of angina relief in those patients refusing restudy at 5 years or more is comparable to that found in those who had late angiography (Table 4). Perioperative and Late MI No difference was noted in the incidence of perioperative MI among the five groups, but a significantly higher rate of late MI was noted in the group showing deterioration (Table 5). Graft Patency versus Ventricular Function and Angina Relief Striking differences were noted in graft patency among the five groups. The groups with normal ventricular function that was unchanged, with improvement resulting in normal LV function, and with improvement but still abnormal LV function all had significantly higher graft patency rates at 5 years after CABG than the group with abnormal ventricular function that did not change or the group that deteriorated (Table 6).

3 549 Tyras et al: Ventricular Function and Native Coronary Circulation Table 2. Postoperative Ventricular Function (35 Years) and Patient Population Characteristics in 181 Patients Ventricular Function Factor No. of patients (YO) No. of graftdpatient Mean age (yr) No. of women (YO) Hypertension (YO) Diabetes (YO) Smoking (YO) Preoperative history of MI (Yo) Cholesterol mg (%) Lipid abnormality IMP- SAME- IMP- SAME- NLV NLV ALV ALV WORSE p Value 40 (22.1) (7.5) 3 (7.5) 34 (85) 15 (37.5) 49 (27.1) (26.5) 14 (28.6) 6 (12.2) 35 (71.4) 13 (26.5) 27 (14.9) l(3.7) 2 (7.4) 3 (11.1) 21 (77.8) 19 (70.4) 30 (16.6) l(3.3) 6 (20) 4 (13.3) 21 (70) 16 (53.3) 35 (19.3) (14.3) 5 (14.3) 6 (17.1) 23 (65.7) 22 (62.9) p < 0.05 p < (30) 14 (28.6) 11 (40.7) 13 (43.3) 15 (42.9) 21 (52.5) 26 (53.1) 15 (55.6) 15 (50) 17 (48.6)... See text IMP = improved from operative state; NLV = normal ventricular function; SAME = unchanged from preoperative state; ALV = abnormal ventricular function; WORSE = deterioration of ventricular function; = not significant; MI = myocardial infarction. Table 3. Ventricular Function versus Angina Relief Current Angina Status (Canadian Heart Association Classification [7]) No. of Patients (%) Death Ventricular after Function Years Cause IMP-NLV 3 (7.5) 6 (15) SAME-NLV (72.3) (77'5)ia 1 (2.1) 7 (14.9) 3 (6.4) 2 (4.3) 2 MI (11, Ca (1) IMP-ALV 10 (37.0) 4 (14.8) 9 (33.3) l(3.7) 3 (11.1) 0 SAME-ALV a 3 (10.3) 7 (24.1) 0 (0) 4 (13.8) 1 MI WORSE 13 (39.4) 9 (27.3) 5 (15.2) 2 (6.1) 4 (12.1) 2 MI (2) apatients free from angina: IMP-NLV and SAME-NLV vs IMP-ALV and SAME-ALV and WORSE, p < Ca = carcinoma; other abbreviations same as for Table 2. Table 4. Current Angina Status in Available Patients Not Undergoing Five-Year Follow-up Study Compared with Those Receiving Late Angiography No. of Patients in Canadian Heart Association Angina Class: Angiography Status Refused angiography (YO)" 121 (58.2) 33 (15.9) 33 (15.9) 10 (4.8) 11 (5.2) Received angiography (YO) 103 (58.5) 20 (11.4) 34 (19.3) 6 (3.4) 13 (7.4) "Includes 11 patients unable to undergo angiography because of severe peripheral vascular disease.

4 ~ ~ 550 The Annals of Thoracic Surgery Vol 27 No 6 June 1979 Table 5. lncidence of Perioperative and Late MI Percent of Patients Ventricular Periop- Function erative MI Late MI IMP-NLV 5.0 SAME-NLV IMP-ALV SAME-ALV WORSE " "p < Abbreviations same as for Table 2. Table 6. Graft Patency at 35 Years versus Ventricular Function Ventricular Function IMP-NLV SAME-NLV IMP-ALV SAME-ALV WORSE "p < Abbeviations same as for Table 2. Graft Patencv Rate (YO) }" 64.5 Also, graft patency decreased stepwise as severity of angina increased. It was 84.2% in those patients with no angina versus 60% in those with Class 3 or 4 angina (p < 0.005). For patients in Class l it was 72.5% and for those in class 2, 66.2%. Completeness of Revascularization The proportion of patients judged to be completely revascularized at operation was essentially the same in.all five groups (Table 7). Also, the percentage of ungrafted vessels with stenosis of 50% or more but less than 100% was the same in all five groups. Progression of Disease in Native Coronary Arteries The progression of atherosclerotic stenosis in coronary arteries receiving bypass grafts was essentially the same in all five groups, ranging from 58.8% in the group showing improvement resulting in normal LV function to 67.4"/0 in the group with normal LV function that was unchanged (Fig 1). Progression of disease in coronary arteries with occluded grafts was no different from those with patent grafts. In ungrafted coronary arteries, progression of disease was significantly lower in the groups with normal LV function (improved, 33.3%, and unchanged, 33.3%) than in those with deterioration of LV function (61.3%) ( p < 0.05). It should also be noted that disease progression in the group showing deterioration was essentially the same in both grafted (61.4%) and ungrafted (61.3%) arteries. There was a significant difference in the incidence of progression of disease in grafted versus ungrafted arteries in those with normal ventricles (improved group, 58.8% vs 33.3% [ p < ; unchanged group, 67.4% vs 33.3% [ p < ). The incidence of significant graft stenosis (diameter reduction of 50% or more) of patent grafts was low, ranging from 2.3% in the group with improvement resulting in normal LV function to 7.9% in the group with improvement but still abnormal LV function. Progression of Disease to Total Occlusion Total occlusion of native coronary arteries that had been patent on preoperative coronary arteriograms was significantly higher in grafted arteries (63.3%) than in those ungrafted (12.2%) (p < 0.001). However, progression to total occlusion was also related to the severity of preoperative stenosis in both grafted and ungrafted arteries. Whereas only 1.2% of ungrafted vessels with preoperative stenosis of less than 50% progressed to total occlusion, 43.8% of ungrafted coronaries progressed from stenosis of 90% or more to total occlusion (Fig 2). In grafted arteries, 48.7% progressed from less than 90% stenosis to total occlusion, while 73.5% with preoperative stenosis of 90% or more were totally occluded at the time of late angiography. Comment Treatment of ischemic heart disease has been directed at symptomatic relief of angina pectoris, increasing longevity, prevention of myocardial infarction, and preservation of ventricular function. This report focuses on ventricular function and progression of

5 551 Tyras et al: Ventricular Function and Native Coronary Circulation Table 7. Completeness of Revascularization Patients No. of Ungrafted Completely No. of No. of Vessels with Ventricular Revascularized Vessels Vessels Stenosis >50% CI(A + B) Function 00) Grafted (A) Ungrafted (B) but <loo% (C) (%I IMP-NLV SAME-NLV IMP-ALV SAME-ALV WORSE Abbreviations same as for Table z 0 Ln 5 % P n - Ln 8 25 E 0 Ungrafled Arteries Grafled Arteries p<0.025 t pto.025 t ~ ~0.05 IMP - NLV SAM - NLV IMP - ALV SAM - ALV WORSE POSTOPERATIVE LV FUNCTION Fig 1. Postoperative left ventricular (LV) functional category and incidence of progression of atherosclerotic disease in ungrafted and grafted coronary arteries. (IMP = improved from preoperative state; NLV = normal ventricular function; SAME = unchanged from preoperative state; ALV = abnormal ventricular function; WORSE = deterioration of ventricular function.) - loo k z t * p<o.oot t pto.025 pt0.025 ~ 2 90% DEGREE OF PREOPERATIVE STENOSIS Fig 2. Incidence of progression of disease to total occlusion as a function of the degree of preoperative coronay artery stenosis in ungrafted and grafted arteries. atherosclerotic disease in the native coronary circulation assessed at least 5 years after CABG in 181 patients. It is noteworthy that almost 50% of these patients have maintained normal LV function for 5 years following CABG. This maintenance of normal LV function is linked to graft patency, with more than 80% of grafts patent at 5 years or more in those patients with normal LV function compared with 64.5% in those with deterioration of LV function (p < 0.05) (see Table 6). These findings are confirmatory of the previous work of Wolf and colleagues [24], which showed that patent coronary artery grafts were associated with maintenance of normal LV function or improvement of regional and global function when preoperative dysfunction was present, but that occluded grafts were often associated with deterioration of LV contractility as measured by segmental wall shortening velocity. The present study provides evidence of the effects of CABG on ventricular function at 5 or more years after operation. It confirms previous observations made early after CABG demonstrating increases in myocardial contractile force [lo, 11, 191. In addition, this study shows the close relationship of graft patency to angina relief, with 84.2% of grafts patent in patients now free of angina compared with 60% patency in those with Class 3 or 4 angina (p < 0.005). This tends to refute the simplistic view that angina relief is frequently related to infarction or a placebo effect rather than to revascularization [I, 41. Analysis of progression of atherosclerosis in

6 552 The Annals of Thoracic Surgery Vol 27 No 6 June 1979 the native coronary circulation following CABG has generally been made on limited numbers of patients or short follow-up periods. Aldridge and Trimble [2] initially reported acceleration of atherosclerosis proximal to a vein graft. Malinow and associates [17] showed in 100 patients studied with angiography less than six months following CABG, that progression of disease to total occlusion occurred in 44% of grafted arteries but in only 6% of ungrafted arteries. They thought this progression was more likely with a patent graft than an occluded graft. Maurer and co-workers 1181 repeated angiography in 121 patients approximately 1 year following CABG and found progression of proximal disease five times more frequent in grafted than in ungrafted arteries. Total occlusion had occurred in 60% of grafted arteries but in only 2% of ungrafted arteries. Better understanding of these preliminary observations is obtained with the perspective of longer follow-up. Seides and colleagues [20], in a limited study of 22 patients who had undergone CABG at least 5 years earlier, noted a 68% incidence of disease progression of 28 grafted vessels at 5 years; but 35% had shown progression at the first recatheterization less than 1 year following CABG. In ungrafted arteries, on the other hand, only 1 of 24 vessels showed progression of disease in an early follow-up angiogram, but 46% had shown significant progression at 5 years. In a more comprehensive study of 108 patients followed for 5 to 7 years after CABG, Bourassa and associates [51 identified progression of atherosclerotic disease at 1 year in 57% of grafted arteries with patent grafts and in 53% with occluded grafts. At 5 years, progression of disease was present in 66% and 57% of arteries with patent or occluded grafts, respectively. In contrast, only 9.5% of ungrafted arteries showed disease progression at 1 year, but 41% had significant increases in atherosclerotic stenosis at 5 to 7 years. Disease progression was more likely with greater degrees of preoperative stenosis, being only 16% in ungrafted arteries with stenosis of 50% or less but as high as 50% in ungrafted arteries with more than 75% stenosis. In grafted arteries with 50 to 75% stenosis, 41% had progression at 5 years; with stenosis of more than 75%, more than 70% showed significant increases in proximal disease. The results of the Montreal group [5, 161 are strikingly similar to those we have described. Among the five groups of postoperative ventricular function, disease progression was noted in grafted arteries in 58.8% to 67.4%. In ungrafted arteries, disease progression ranged from 33.3% in the groups with normal LV function to 61.3% in those with deterioration of ventricular function, with an overall rate of 42% in the 181 patients studied. Progression of disease in grafted arteries appears to occur predominantly in the first year following operation. This may be a hemodynamic phenomenon resulting from decreased blood flow through a stenotic coronary artery because of a patent vein graft. Experimental work by Furuse and co-workers [8] and Kakos and associates [141 are supportive of this concept. It seems unlikely that operative manipulation of the coronary artery is a major factor since only 8.4% of grafted arteries had stenosis at the graft site, and 2.4% had progression distal to the graft in the Montreal patients studied more than 5 years following CABG [51. The data presented suggest that maintenance or restoration of normal ventricular function after CABG is related to graft patency and to the degree of progression of atherosclerotic disease in ungrafted coronary arteries. Our finding that the group of patients with deterioration of LV function showed a level of disease progression in ungrufted arteries (61.3%) equal to that found in grafted arteries could indicate that low graft patency, occurrence of late myocardial infarction, and deterioration of ventricular function in this group might all be related to intrinsic acceleration of coronary atherosclerosis unrelated to operative intervention. The finding that almost 50% of the patients reviewed more than 5 years following CABG still maintain normal LV function is an encouraging, objective indication of long-term benefit from operative therapy of ischemic heart disease. Long-term studies of operatively treated patients evaluated with ventriculography and angiography more than 5 years after CABG can best provide meaningful data about

7 553 Tyras et al: Ventricular Function and Native Coronary Circulation graft patency, angina relief, myocardial infarction, ventricular function, and native coronary arterial disease progression. References 1. Achuff SC, Griffith LSC, Conti CR, et al: The "angina-producing" myocardial segment: an approach to the interpretation of results of coronary bypass surgery. Am J Cardiol 36:723, Aldridge HE, Trimble AS: Progression of proximal coronary artery lesions to total occlusion after aorto-coronary saphenous vein bypass grafting. J Thorac Cardiovasc Surg 62:7, Bamer HB, Laks H, Codd JE, et al: Cold blood as the vehicle for potassium cardioplegia. Ann Thorac Surg (in press, 1979) 4. Benchimol A, dos Santos A, Desser KB: Relief of angina pectoris in patients with occluded coronary bypass grafts. Am J Med 60:339, Bourassa MG, Lesperance J, Corbara F, et al: Progression of obstructive coronary artery disease 5 to 7 years after aortocoronary bypass surgery. Circulation 58:Suppl 1:100, Cameron A, Kemp HG Jr, Shimomura S, et al: Coronary artery bypass surgery: long-term follow-up. NY State J Med 77:27, Campeau L: Grading of angina pectoris (letter to the editor). Circulation 54:522, Fumse A, Klopp EH, Brawley RK, et al: Hemodynamics of aorta-to-coronary artery bypass: experimental and analytical studies. Ann Thorac Surg 14:282, Griffith LSC, Achuff SC, Conti CR, et al: Changes - in intrinsic coronary circulation and segmental ventricular motion after saphenous vein coronary bypass graft surgery. N Engl J Med 288:589, Hairston P, Newman WH, Daniel1 HB: Myocardial contractile force as influenced by direct coronary surgery. Ann Thorac Surg 15:364, Johnson PE Jr, Briggs H, Ishikawa K, et al: Evaluation of the immediate effect of aortocoronary saphenous vein bypass surgery on myocardial contractility. Chest 66:50, Judkins MI': Percutaneous transfemoral selective coronary arteriography. Radio1 Clin North Am 6:467, Kaiser GC, Bamer HB, Tyras DH, et al: Myocardial revascularization: a rebuttal of the cooperative study. Ann Surg 188:331, Kakos GS, Oldham HN, Dixon SH Jr, et al: Coronary artery hemodynamics after aortocoronary vein bypass: an experimental evaluation. J Thorac Cardiovasc Surg , Lawrie GM, Morris GC Jr, Howell JF, et al: Results of coronary bypass more than 5 years after operation in 434 patients: clinical, treadmill exercise and angiographic correlations. Am J Cardiol 40:665, Lesperance J, Bourassa MG, Saltiel J, et al: Angiographic changes in aortocoronary vein grafts: lack of progression beyond the first year. Circulation 48:663, Malinow MR, Kremkau EL, Kloster FE, et al: Occlusion of coronary arteries after vein bypass. Circulation 47:1211, Maurer BJ, Oberman A, Holt JH Jr, et al: Changes in grafted and nongrafted coronary arteries following saphenous vein bypass grafting. Circulation 50:293, Moran SV, Tarazi RC, Urzua JU, et al: Effects of aorto-coronary bypass on myocardial contractility. J Thorac Cardiovasc Surg 65:335, Seides SF, Borer JS, Kent KM, et al: Long-term anatomic fate of coronary-artery bypass grafts and functional status of patients five years after operation. N Engl J Med 298:1214, Sheldon WC, Rincon G, Pichard AD, et al: Surgical treatment of coronary artery disease: pure graft operations, with a study of 741 patients followed 3-7 years. Prog Cardiovasc Dis 18:237, Tyras DH, Barner HB, Kaiser GC, et al: Myocardial revascularization in women. Ann Thorac Surg 25:449, Tyras DH, Bamer HB, Kaiser GC, et al: Longterm benefits of myocardial revascularization. Presented at the 27th Annual Meeting of the American College of Cardiology, Anaheim, CA, March 6-9, Wolf NM, Kreulen TH, Bove AA, et al: Left ventricular function following coronary bypass surgery. Circulation 58:63, 1978