- Email: [email protected]
Angina following myocardial revascularization
Angina following myocardial revascularization Does time of recurrence predict etiology and influence results of operation? To assess the operative mortality and long-term results in patients undergoing repeat revascularization for recurrent angina, we analyzed 48 consecutive patients operated upon at New York University Medical Center between 1970 and 1978. Between January, 1970, and July, 1973, 15 patients underwent repeat rcvascularization with five operative deaths (33 percent). Thirty-three patients underwent similar operations from July, 1973, to July, 1978, with only one operative death (3 percent). Technical factors and improved methods of myocardial protection during the operation directly influence this decrease in operative mortality rate. The indication for reoperation It'as incapacitating angina developing within 2 months of the initial operation in 18 patients (early failures) and after more than 2 months in 30 patients (late failures). The early failures were most commonly attributed to technical factors (33 percent) and graft occlusion by exuberant pericardial scarring (33 percent). The late failures were commonly related to the development of new native coronary lesions (47 percent) and selection of an incorrect site for distal anastomoses (23 percent). The prognostic and therapeutic implications of these findings will be discussed in detail. Angina was abolished or significantly decreased in 90 percent of the survivors, and there were only two late deaths occurring 18 and 20 months postoperatively. These data indicate that patients undergoing repeat myocardial revascularization can be operated upon with low operative mortality rates and symptomatic improvement comparable to that of patients undergoing coronary artery bypass for the first time.
Alfred T. Culliford, M.D., Robert W. Girdwood, M.D., O. Wayne Isom, M.D., Kenneth R. Krauss, M.D., and Frank C. Spencer, M.D. New York, N. Y.
F
or the past decade surgical therapy for coronary artery disease has been a major area of clinical investigation at New York University Medical Center. The safety and effectiveness of coronary artery bypass surgery in relieving incapacitating angina refractory to medical management has now been unequivocably established. Some patients will require a second myocardial revascularization because of the appearance of severely disabling angina after the original operation. The experience at New York University Medical Center with this procedure will be thoroughly reviewed. Factors noted to be responsible for the initial surgical failure, From the Departments of Surgery and Medicine, New York University Medical Center, 560 First Ave., New York, N. Y. 10016.
Received for publication July 18, 1978. Accepted for publication Feb. 27, 1979. Address for reprints: Alfred T. Cuiliford, M.D., Department of Surgery, New York University Medical Center, 560 First Ave., New York, N. Y. 10016.
operative morbidity and mortality, and their relationship to the time interval between the original operation and the appearance of recurrent angina will be discussed.
Patients From November, 1970, to August, 1978, 48 consecutive patients have undergone a second procedure for myocardial revascularization (Table I). Ages range from 35 to 70 years with a mean age of 53.4 years. The male-to-female ratio was 3.2 to 1. All patients received direct coronary artery bypass grafting during initial revascularization. There were four single, 22 double, and 22 triple bypasses employing the internal mammary artery or autologous saphenous vein.
Definition of patient groups When the time of onset of recurrent angina is studied in relation to original revascularization, two distinct groups of patients are noted (Fig. 1). Group I, consisting of 18 patients, began having severe and incapacitat-
0022-5223/79/060889+07$00.70/0 © 1979 The C. V. Mosby Co.
889
The Journal of Thoracic and Cardiovascular Surgery
8 90 Culliford et al.
20
5
2
24
12
30
48
60
Time After Original Revascularizing Surgery (Months)
Fig. 1. Onset of recurrent angina after original operation in 51 patients undergoing a second procedure for myocardial revascularization.
Table I. Forty-eight patients undergoing repeated myocardial revascularization * Original bypass operation
No. of patients
Single bypass graft Double bypass grafts Triple or more bypass grafts
22 22
4
'The patients were 35 to 70 years of age, mean 53.4 years. The male-tofemale ratio was 3.2 to I.
ing angina within the first 2 postoperative months.
Group II, consisting of 30 patients, experienced recurrent symptoms from 2 months to 5 years after the original operation. Each group, categorized according to the time interval from the original operation to the appearance of severe symptoms, was studied to identify the causative factors responsible for recurrent angina as well as the results of reoperation. In these two areas, particularly the former, striking differences were revealed.
Methods Technique of resternotomy. Forty-seven patients required resternotomy for secondary revascularization. One patient, who required only a circumflex graft, was operated upon through a left lateral thoracotomy incision.
The details of operation through a previous sternotomy are of critical importance and are described in a separate publication. Technique of cardiopulmonary bypass. Operations were done with the aid of cardiopulmonary bypass. Before heparinization the incision was checked thoroughly for hemostasis. The ascending aorta was cannulated in all patients. In some patients, in whom difficulty with sternal opening was anticipated, the femoral artery was exposed. Fortunately, the urgent cannulation of this vessel was not required and all resternotomies proceeded uneventfully. A disposable bubble oxygenator primed with electrolyte solution was routinely used. Flow rates were 2.0 to 2.5 L. per meter squared per minute and were adjusted to maintain perfusion pressures equal to the mean arterial pressure.
Technique of cardiac arrest and coronary anastomosis. Distal coronary anastomoses were constructed with the aorta cross-clamped after electrically induced fibrillation. Systemic hypothermia (25 0 C.) was used during these periods of ischemia. Iced normal saline solution was used for topical hypothermia in most patients but not all. Optical magnification with four-power operating loupes was used during dissection of coronary vessels and creation of the anastomoses. The distal reconstruction was carried out with 6-0 or 7-0 polypropylene su-
Volume 77 Number 6 June, 1979
Angina following myocardial revascularization
Table II. Comparison ofpreoperative status, findings at coronary angiography, and initial operation for patients with early and late recurrent angina
Table III. Comparison of coronary angiography, operation performed, and postoperative course for patients with early and late recurrent angina
I Group 1* I Group
lit
18 4(22%)
30 1(3%)
Total patients
Findings at catheterization: Obstructed vessels per patient LVEDP (mm. Hg)
2.5 12
2.4 14.7
Procedures performed: Bypasses/patient Mean graft flow: Under 50 ml./min.
2.4 55%
2.3 50%
Findings at recath: Grafts closed Occluded grafts patient Patients with new lesions LVEDP (mm. Hg)
Postop status: Postop. myocardial infarction Reexploration for bleeding
~ No.
Preop. status: Total patients Preinfarction angina
17% 0
3.5% 7%
Legend: LVEDP, Left ventricular end-diastolic pressure. "Recurrent angina within the first 2 postoperative months.
tRecurrent angina after the second postoperative month.
ture. If the vessels were small or extensively coated with plaque, interrupted sutures were used for proximal and distal ends of the anastomoses.
Results Preoperative status and primary myocardial reo vascularization (Table II). Preoperative status and findings at catheterization. Of 18 patients exhibiting recurrent symptoms during the first 2 postoperative months (Group I), four (22 percent) had preinfarction angina, whereas only one patient (3 percent) among 30 patients with the late onset of recurrent angina (Group II) had preinfarction angina. Initial cardiac catheterization revealed a similar number of major coronary artery obstructions: 2.5 obstructed vessels per patient in Group I and 2.4 obstructed vessels per patient in Group II. Left ventricular function, as assessed by ventriculography and measurement of left ventricular end-diastolic pressure, was virtually the same for the two groups. Initial procedures performed and postoperative status. In Group I an average of 2.4 bypasses was performed in each patient, which compared favorably to the findings at catheterization, i.e., 2.5 obstructed vessels per patient. The patients in Group II received 2.3 bypass grafts per patient, compared to the 2.4 obstructed vessels noted per patient. When flow rates were measured in the operating room at the completion of revascularization, 50 to 55 percent of the patients in both groups had mean flows under 50 ml. per minute.
Procedures performed: Total grafts inserted Bypasses/patient Occluded grafts redone New lesions bypassed Mean graft flow: Under 50 ml./min. Technique of repair: IMA or vein bypass vein patch Postop. status: Postop. myocardial infarction Re-exploration for bleeding
%
18
34 1.9 1 12.4
79 5
891
No.
%
30 41 1.4 14 15.1
60 47
31 1.7 30 88 I 100 50
46 1.5 34 12
28 2
93 7
53 5
91 9
2 2
II II
3 2
11 7
82 86 40
Legend: LVEDP. Left ventricular end-diastolic pressure. IMA, Internal mammary artery. "Recurrent angina within the first 2 postoperative months. tRecurrent angina after the second postoperative month.
Postoperatively, 17 percent of the early-onset group and 3.5 percent of the late-onset group had electrocardiographic and/or enzymatic evidence of an acute myocardial infarction. The incidence of postoperative hemorrhage necessitating re-exploration was the same in both groups.
Results in recatheterization, operation performed, and postoperative course in 48 patients developing recurrent angina (Table III). Findings at recatheterization. Several striking differences were noted when patients with early (Group I) and late (Group II) appearance of recurrent angina were studied. In Group I, 79 percent of the original grafts had closed, as compared to 60 percent in Group II. Fourteen patients (47 percent) in the late-onset group were noted to have new native coronary artery lesions compared to only one patient (5 percent) in the earlyonset group. Ventricular function as assessed by ventriculography and measurement of left ventricular end-diastolic pressure was not appreciably different between the two groups. Interestingly, when ventricular function was compared to preoperative function, no significant difference was noted in either group.
892
The Journal of Thoracic and Cardiovascular Surgery
Culliford et al.
Table IV. Factors responsible for the development of recurrent angina early (Group /) and late (Group II) after initial revascularization Group 1* Group lIt Cause of original surgical failure
No.1 %
6 4 6
Occlusion by dense pericardial scar Incorrect site chosen for distal anastomosis Technical failure of anastomoses Progressive obliteration of graft lumen Appearance of new native coronary lesions
No·1 %
33 22 33
3 7 4
10 23 13
I
6
2
7
I
6
14
47
'Recurrent angina within the first 2 postoperative months. tRecurrent anging after the second postoperative month.
Table V. Operative mortality rate for 48 patients undergoing a second procedure for myocardial revascularization
Time period
Operative deaths
Patients having second revasc. procedure
January, 1970, to July, 1973 July, 1973, to July, 1977
No.
15 33
1%
5
33 3
I
Table VI. Long-term follow-up on patients surviving repeated myocardial revascularization Group 1* (/5 survivors)
I
Group lIt (27 survivors)
I
%
No.
13
87
23
85
2
13
4
15
Status of angina
No.
Angina absent or significantly decreased Angina present
%
'Recurrent angina within the first 2 postoperative months. tRecurrent angina after the second postoperative month.
Secondary procedures performed and postoperative status. It was possible to reconstruct 30 of the 34 (88 percent) occluded grafts in Group I; four (12 percent) grafts were not revised (two because the vessels were not technically satisfactory, one vessel could not be found in dense scar, and one patient with occluded grafts had a cardiac arrest at the induction of anesthesia and could not be resuscitated). In Group 11,34 of the 41 occluded grafts (82 percent) were reconstructed. Seven grafts were not revised (three vessels were too small for grafting, two vessels were diffusely diseased, and two vessels could not be located because of dense scarring).
The internal mammary artery or saphenous vein was used in the majority of patients. A vein patch angioplasty technique was particularly useful in patients with widely patent vein grafts and anastomotic strictures caused by progression of disease, epicardial scar stricture, or technical factors. A total of 13 patients had new nati ve coronary artery lesions bypassed. In Group I only one of the patients (5 percent) was noted to have a new native lesion. In contrast, 12 of the 30 patients in Group II (47 percent) had new native lesions and nearly all were successfully bypassed. The occurrence of postoperative infarction was the same for Groups I and II. Four patients in the study (8 percent) required re-exploration for postoperati ve hemorrhage.
Factors responsible for the development of recurrent angina early (Group I) and late (Group II) after initial revascularization (Table I V). Factors responsible for failure after original revascularization were derived by review of the operative notes, pathology reports, and preoperative angiograms. In most of the patients it was possible to assign one predominant factor responsible for the recurrence of severe angina pectoris. Occlusion by dense pericardial scar. Six patients (33 percent) in the early-onset group and three patients (10 percent) in the late-onset group were found to have exuberant and dense pericardial scar obstructing previously inserted grafts. Incorrect site chosen for distal anastomoses. Placement of original bypass grafts into nonobstructed distal vessels or proximal to stenotic lesions constituted another reason for failure. This error appeared in four patients (22 percent) of the early-onset group and seven patients (23 percent) in the late-onset group. Technical failures of anastomoses. Improper anastomotic technique was a common cause of failure. Commonly, this involved improper placement of sutures at the toe or heel (or both) of the anastomoses, resulting in ultimate occlusion of the graft. Occasionally, the coronary vessel had been opened either obliquely or eccentrically, so that the anastomosis had been difficult and technically awkward to construct. As expected, this factor was noted in six patients (33 percent) of the early-onset group and in only four patients (13 percent) in the late-onset group. Progressive obliteration of the graft. Some saphenous vein grafts were noted to be obstructed by dense myxoid proliferation. This uncommon complication occurred in a total of three patients for both groups. Appearance of new native coronary lesions. Changes
Volume 77 Number 6 June, 1979
in the native coronary circulation were detected by comparing the coronary angiogram before each revascularization procedure. New lesions were noted in 14 patients (47 percent) of the late-onset group and only one patient (6 percent) of the early-onset group. Operative deaths (Table V). Deaths before discharge of the patient from the hospital or within the first postoperative month were considered as operative deaths. A total of six deaths fulfilled these requirements (12.5 percent). When the time period of this study, January, 1970, to August, 1978, was divided into an initial phase-January, 1970, to July, 1973-and a late phase-August, 1973, to July, 1978-a marked change in operative mortality rate was seen. During the first half of this study, five of 15 patients died, a mortality rate of 33 percent. During the last half of the study, only one of 33 patients died, for a mortality rate of 3 percent. Four deaths were due to myocardial infarctionthree occurring in the operating room and one during the second postoperative month. One patient died of a massive pulmonary embolus and one patient died of sepsis from pulmonary complications. Improved methods in anesthetic technique, intraoperative monitoring, and myocardial preservation during coronary bypass grafts are responsible for the significant and gratifying decrease in the operative mortality rate. Late results (Table VI). Follow-up data on all 42 surviving patients were obtained during outpatient follow-up and phone conversations with the referring physician. Status of angina. Thirty-six of the 42 patients (86 percent) were experiencing no angina or rare episodes of angina associated with extreme exertion. When assessed on the basis of the interval between original operation and the reappearance of angina, no appreciable difference was noted. Late deaths. There were two late deaths. One patient died suddenly at home 2'12 years postoperatively of a presumed myocardial infarction. The second patient died of sepsis 7 months following the operation. Discussion From January, 1970, to July, 1978, more than 1,500 patients with severe angina refractory to medical management underwent direct myocardial revascularization at New York University Medical Center. Gratifying symptomatic improvement was noted in 92 percent of the patients with a minimal operative risk. Long-term analysis of our data confirms the impression that coronary artery bypass grafting increases longevity for pa-
Angina following myocardial revascularization
893
tients with multiple coronary obstructions." This observation has been confirmed by others.t' During the same time period 48 patients (3.2 percent) required a second direct revascularizing procedure for the reappearance of incapacitating angina. A careful analysis of these patients has been conducted to answer two fundamental questions: First, what were the specific factors responsible for initial surgical failure? Second, who will benefit from a second procedure and at what risk? Patients who begin having recurrent angina within 2 months of operation (Group I) and those with the later onset of angina (Group II) were noted to differ in several significant ways. Since the time of onset of recurrent pain is the single objective parameter available when patients are initially evaluated, this means a classification is practical and provides valuable prognostic information. Factors predisposing to initial surgical failure. Preoperative factors prior to initial revascularization. Adequate and complete assessment of native coronary circulation is a prerequisite for properly planned revascularization. The coronary vessels should be visualized in multiple radiographic exposures to determine the extent and location of atherosclerotic plaques. A careful study is required of vessels that are totally obstructed during direct proximal injection but that are visualized during retrograde filling. These vessels are frequently of large caliber and can be overlooked easily because they fill with small quantities of dye during angiography. Major vessels that are more than 70 percent obstructed should be bypassed. Wukasch and associates" recently have suggested the importance of revascularizing vessels which are only 50 to 70 percent obstructed at the time of initial operation. This suggestion is based on their analysis of 41 patients requiring repeated myocardial revascularization. During initial revascularization 10 arteries with 50 to 70 percent obstructions were not bypassed. Of these, nine had progressed to significant stenosis within 6 months to 4 years after initial operation, and in five patients they were the only reason for reoperation. This concept requires additional documentation before it can be recommended, but the data reveal a striking progression of disease in a relatively short period of time. Intraoperative factors during initial revascularization. Two intraoperative factors determine to a large extent the success of coronary bypass grafting: surgical techniques and completeness of revascularization. Vein graft failure was attributed to purely technical factors in II patients-six in Group I (33 percent) and four in Group II (13 percent). Distal coronary anastomoses were placed proximal to significantly obstruct-
894
Culliford et al.
ed lesions or in the wrong vessel in 11 patients-four in Group I (22 percent) and seven in Group II (23 percent). These factors underscore the importance of precise surgical technique during coronary anastomosis. Coronary arteriotomy should be done distal to significant obstruction in a region where the vessel is uninvolved with atherosclerosis. This site is apparent when the telltale "blue line," representing the lumen distended with blood, is noted in the coronary artery. Following arteriotomy the vessel is sized and gently probed to assure an unobstructed distal lumen. Anastomosis is carried out with a 6-0 to 7-0 polypropylene suture. Optical magnification and the high-intensity head lamp enhance visualization of the fine details of coronary anatomy. Graft obstruction by neointimal hyperplasia with ultimate closure of the lumen was noted in only three patients. This gratifying finding is attributable to gentle dissection of the saphenous vein, distention of the vessel with cold heparinized blood prior to insertion, and minimal direct handling of the intima and adventitia during preparation of the vein graft and actual anastomosis. The completeness of revascularization has been studied carefully in this series of patients. At the time of initial operation each patient received an average of 2.4 grafts compared to the 2.5 obstructed vessels noted at angiography, for a revascularization rate of 96 percent. As a result of this policy, no patient in this series had recurrent and disabling angina attributable to incomplete initial revascularization. This high revascularization rate is in marked contrast to other reported experiences. Johnson and co-workers" noted that of the 39 patients requiring a second operation, only three had triple bypass grafts initially. This group now frequently bypasses vessels which are 40 to 50 percent obstructed at the time that more severe lesions are being bypassed in an effort to achieve complete revascularization. In a series of 219 cases recently reported from the Cleveland Clinic, incomplete initial revascularization was responsible for 18 percent of the secondary operations for angina." Similar experiences have been reported by Winkle and co-workers ,8 with 21 reoperations. In this series 50 percent of the patients were initially treated with single bypass grafting and 40 percent received double bypass grafting. Nine critical lesions were not bypassed at the original operation, all of which were in the circumflex system. The authors noted that patients recei ving "total surgical correction" of obstructed coronary vessels after initial and secondary procedures have significantly greater improvement
The Journal of Thoracic and Cardiovascular Surgery
in the level of angina than those "not totally corrected. " Postoperative factors predisposing to initial surgical failure. Graft occlusion was attributed to the presence of dense pericardial scar in 33 percent of Group I and 10 percent of Group II patients. These findings were noted in patients who exhibited a significant postpericardiotomy reaction following the initial operation. Typically, this was manifested by fever, malaise, prolonged pleuropericardial pain, and a persistent rub. Significant leukocytosis and an abnormal number of lymphocytes were commonly noted during the immediate postoperative course. These findings and symptoms prompt a thorough investigation to exclude other sources of infection. Once they are excluded, therapy with either acetylsalicylic acid or a short course of steroids is instituted. Similar observations and conclusions have been reported by Oglietti ," Adam, 10 and their colleagues. Significant factors during repeated revascularization predisposing to successful outcomes. In the past, resternotomy itself has been a hazardous and sometimes fatal operation. Following the operative guidelines briefly described, resternotomy was performed in the 48 consecutive patients without surgical misadventure necessitating the premature institution of cardiopulmonary bypass. Re-exploration for bleeding was nevertheless necessary in four patients or 8 percent, which is nearly eight times the re-exploration rate for primary sternotomy. Technical revision of occluded grafts was possible in 93 percent of the cases. This was usually accomplished with a new saphenous vein or internal mammary graft; however, 7 percent of occluded grafts were directly revised by vein patch angioplasty. The greater frequency of new native coronary artery lesions in Group II (47 percent) is not surprising and has been noted in other series. 8. 9. 11 Fortunately, all new lesions in both groups could be technically bypassed. The importance of a complete secondary revascularization is of paramount significance in achieving gratifying long-term results. Operative mortality rate and long-term results. Although the over-all mortality rate for the entire series is 12 percent, only one death has occurred in the last 33 consecutive patients (3 percent). As experience has accumulated with this procedure, simultaneous refinements in anesthetic management, intraoperative monitoring, and improved methods of myocardial preservation have enhanced the safety of a second revascularization.
Volume 77 Number 6
Angina following myocardial revascularization
895
June. 1979
Significant and impressive reduction or ablation of angina was noted in long-term survivors. The over-all reduction or ablation of angina is attributable to an aggressive policy of revascularization in patients with failed grafts and new coronary artery lesions. These results compare favorably with other series in which the reported frequency of significant postoperative improvement has ranged from 47 to 90 percent. 5.8.9. 11-13 From these data, it is apparent that repeated myocardial revascularization can be carried out with a negligible operative risk and offers sustained symptomatic relief in 80 to 90 percent of patients. The time of onset of angina after the initial operation is significant, since it is indicative of the cause of recurrent pain and to a degree predicts the outcome. REFERENCES Isom OW, Spencer FC, Glassman E, Cunningham IN, Teiko P. Reed GE, Boyd AD: Does coronary bypass increase longevity? J THORAC CARDIOVASC SURG 70:28-37, 1978 2 Stiles QR, Lindesmith GG, Tuckes BL, Hughes RK, Meyer BW: Long-term follow-up of patients with coronary artery grafts. Circulation 54:Suppl 3:32, 1976 3 Sheldon WC, Loop FD: Direct myocardial revascularization-1976. Cleve Clin Q 43:97-108, 1976 4 Tecklenberg PL, Alderman EL, Miller DC, Shumway NE, Harrison DC: Changes in survival and symptom relief in a longitudinal study of patients after bypass surgery. Circulation 51, 52:Suppl 1:98, 1975
5 Wukasch DC, Toscano M, Cooley DA, Reul GJ, Sandford FM, Kyger ER, Hallman GL: Reoperation following direct myocardial revascularization. Circulation 56:Suppl 2:3, 1977 6 Johnson WD, Hoffman JF Jr, Flemma RJ, Tector AJ: Secondary surgical procedure for myocardial revascularization J THORAC CARDIOVASC SURG 64:523-529, 1972 7 lrarrazaval MJ, Cosgrove DM, Loop FD, Ennix CL Jr, Groves LK, Taylor PC: Reoperation for myocardial revascularization. J THORAC CARDIOVASC SURG 73: 181188, 1977 8 Winkle RA, Alderman EL, Shumway NE, Harrison DC: Results of reoperation for unsuccessful coronary artery bypass surgery. Circulation 51:Suppl 1:62, 1975 9 Oglietti J, Angelini P, Leachman RD, Cooley DA: Myocardial revascularization. Early and late results after reoperation. J THORAC CARDIOVASC SURG 71:736-740, 1976 10 Adam M, Geisler GF, Lambert CJ, Mitchel BF: Reoperation following clinical failure of aorta-to-coronary artery bypass vein grafts. Ann Thorac Surg 14:272-28 I, 1972 II Thomas CS, Alford WC, Burrus GR, Frist RA, Stoney WS: Results of reoperation for failed aortocoronary bypass grafts. Arch Surg 111:1210-1213, 1976 12 Norwood WI, Cohn LH, Collins 11 Jr: Results of reoperation for recurrent angina pectoris. Ann Thorac Surg 23:9-13, 1977 13 Loop FD, Carabajal NR, Taylor PC, Irarrazaval MJ: Internal mammary artery bypass grafts in reoperative myocardial revascularization. Am J Cardiol 37:890-895, 1976
Alfred T. Culliford, M.D., Robert W. Girdwood, M.D., O. Wayne Isom, M.D., Kenneth R. Krauss, M.D., and Frank C. Spencer, M.D. New York, N. Y.
F
or the past decade surgical therapy for coronary artery disease has been a major area of clinical investigation at New York University Medical Center. The safety and effectiveness of coronary artery bypass surgery in relieving incapacitating angina refractory to medical management has now been unequivocably established. Some patients will require a second myocardial revascularization because of the appearance of severely disabling angina after the original operation. The experience at New York University Medical Center with this procedure will be thoroughly reviewed. Factors noted to be responsible for the initial surgical failure, From the Departments of Surgery and Medicine, New York University Medical Center, 560 First Ave., New York, N. Y. 10016.
Received for publication July 18, 1978. Accepted for publication Feb. 27, 1979. Address for reprints: Alfred T. Cuiliford, M.D., Department of Surgery, New York University Medical Center, 560 First Ave., New York, N. Y. 10016.
operative morbidity and mortality, and their relationship to the time interval between the original operation and the appearance of recurrent angina will be discussed.
Patients From November, 1970, to August, 1978, 48 consecutive patients have undergone a second procedure for myocardial revascularization (Table I). Ages range from 35 to 70 years with a mean age of 53.4 years. The male-to-female ratio was 3.2 to 1. All patients received direct coronary artery bypass grafting during initial revascularization. There were four single, 22 double, and 22 triple bypasses employing the internal mammary artery or autologous saphenous vein.
Definition of patient groups When the time of onset of recurrent angina is studied in relation to original revascularization, two distinct groups of patients are noted (Fig. 1). Group I, consisting of 18 patients, began having severe and incapacitat-
0022-5223/79/060889+07$00.70/0 © 1979 The C. V. Mosby Co.
889
The Journal of Thoracic and Cardiovascular Surgery
8 90 Culliford et al.
20
5
2
24
12
30
48
60
Time After Original Revascularizing Surgery (Months)
Fig. 1. Onset of recurrent angina after original operation in 51 patients undergoing a second procedure for myocardial revascularization.
Table I. Forty-eight patients undergoing repeated myocardial revascularization * Original bypass operation
No. of patients
Single bypass graft Double bypass grafts Triple or more bypass grafts
22 22
4
'The patients were 35 to 70 years of age, mean 53.4 years. The male-tofemale ratio was 3.2 to I.
ing angina within the first 2 postoperative months.
Group II, consisting of 30 patients, experienced recurrent symptoms from 2 months to 5 years after the original operation. Each group, categorized according to the time interval from the original operation to the appearance of severe symptoms, was studied to identify the causative factors responsible for recurrent angina as well as the results of reoperation. In these two areas, particularly the former, striking differences were revealed.
Methods Technique of resternotomy. Forty-seven patients required resternotomy for secondary revascularization. One patient, who required only a circumflex graft, was operated upon through a left lateral thoracotomy incision.
The details of operation through a previous sternotomy are of critical importance and are described in a separate publication. Technique of cardiopulmonary bypass. Operations were done with the aid of cardiopulmonary bypass. Before heparinization the incision was checked thoroughly for hemostasis. The ascending aorta was cannulated in all patients. In some patients, in whom difficulty with sternal opening was anticipated, the femoral artery was exposed. Fortunately, the urgent cannulation of this vessel was not required and all resternotomies proceeded uneventfully. A disposable bubble oxygenator primed with electrolyte solution was routinely used. Flow rates were 2.0 to 2.5 L. per meter squared per minute and were adjusted to maintain perfusion pressures equal to the mean arterial pressure.
Technique of cardiac arrest and coronary anastomosis. Distal coronary anastomoses were constructed with the aorta cross-clamped after electrically induced fibrillation. Systemic hypothermia (25 0 C.) was used during these periods of ischemia. Iced normal saline solution was used for topical hypothermia in most patients but not all. Optical magnification with four-power operating loupes was used during dissection of coronary vessels and creation of the anastomoses. The distal reconstruction was carried out with 6-0 or 7-0 polypropylene su-
Volume 77 Number 6 June, 1979
Angina following myocardial revascularization
Table II. Comparison ofpreoperative status, findings at coronary angiography, and initial operation for patients with early and late recurrent angina
Table III. Comparison of coronary angiography, operation performed, and postoperative course for patients with early and late recurrent angina
I Group 1* I Group
lit
18 4(22%)
30 1(3%)
Total patients
Findings at catheterization: Obstructed vessels per patient LVEDP (mm. Hg)
2.5 12
2.4 14.7
Procedures performed: Bypasses/patient Mean graft flow: Under 50 ml./min.
2.4 55%
2.3 50%
Findings at recath: Grafts closed Occluded grafts patient Patients with new lesions LVEDP (mm. Hg)
Postop status: Postop. myocardial infarction Reexploration for bleeding
~ No.
Preop. status: Total patients Preinfarction angina
17% 0
3.5% 7%
Legend: LVEDP, Left ventricular end-diastolic pressure. "Recurrent angina within the first 2 postoperative months.
tRecurrent angina after the second postoperative month.
ture. If the vessels were small or extensively coated with plaque, interrupted sutures were used for proximal and distal ends of the anastomoses.
Results Preoperative status and primary myocardial reo vascularization (Table II). Preoperative status and findings at catheterization. Of 18 patients exhibiting recurrent symptoms during the first 2 postoperative months (Group I), four (22 percent) had preinfarction angina, whereas only one patient (3 percent) among 30 patients with the late onset of recurrent angina (Group II) had preinfarction angina. Initial cardiac catheterization revealed a similar number of major coronary artery obstructions: 2.5 obstructed vessels per patient in Group I and 2.4 obstructed vessels per patient in Group II. Left ventricular function, as assessed by ventriculography and measurement of left ventricular end-diastolic pressure, was virtually the same for the two groups. Initial procedures performed and postoperative status. In Group I an average of 2.4 bypasses was performed in each patient, which compared favorably to the findings at catheterization, i.e., 2.5 obstructed vessels per patient. The patients in Group II received 2.3 bypass grafts per patient, compared to the 2.4 obstructed vessels noted per patient. When flow rates were measured in the operating room at the completion of revascularization, 50 to 55 percent of the patients in both groups had mean flows under 50 ml. per minute.
Procedures performed: Total grafts inserted Bypasses/patient Occluded grafts redone New lesions bypassed Mean graft flow: Under 50 ml./min. Technique of repair: IMA or vein bypass vein patch Postop. status: Postop. myocardial infarction Re-exploration for bleeding
%
18
34 1.9 1 12.4
79 5
891
No.
%
30 41 1.4 14 15.1
60 47
31 1.7 30 88 I 100 50
46 1.5 34 12
28 2
93 7
53 5
91 9
2 2
II II
3 2
11 7
82 86 40
Legend: LVEDP. Left ventricular end-diastolic pressure. IMA, Internal mammary artery. "Recurrent angina within the first 2 postoperative months. tRecurrent angina after the second postoperative month.
Postoperatively, 17 percent of the early-onset group and 3.5 percent of the late-onset group had electrocardiographic and/or enzymatic evidence of an acute myocardial infarction. The incidence of postoperative hemorrhage necessitating re-exploration was the same in both groups.
Results in recatheterization, operation performed, and postoperative course in 48 patients developing recurrent angina (Table III). Findings at recatheterization. Several striking differences were noted when patients with early (Group I) and late (Group II) appearance of recurrent angina were studied. In Group I, 79 percent of the original grafts had closed, as compared to 60 percent in Group II. Fourteen patients (47 percent) in the late-onset group were noted to have new native coronary artery lesions compared to only one patient (5 percent) in the earlyonset group. Ventricular function as assessed by ventriculography and measurement of left ventricular end-diastolic pressure was not appreciably different between the two groups. Interestingly, when ventricular function was compared to preoperative function, no significant difference was noted in either group.
892
The Journal of Thoracic and Cardiovascular Surgery
Culliford et al.
Table IV. Factors responsible for the development of recurrent angina early (Group /) and late (Group II) after initial revascularization Group 1* Group lIt Cause of original surgical failure
No.1 %
6 4 6
Occlusion by dense pericardial scar Incorrect site chosen for distal anastomosis Technical failure of anastomoses Progressive obliteration of graft lumen Appearance of new native coronary lesions
No·1 %
33 22 33
3 7 4
10 23 13
I
6
2
7
I
6
14
47
'Recurrent angina within the first 2 postoperative months. tRecurrent anging after the second postoperative month.
Table V. Operative mortality rate for 48 patients undergoing a second procedure for myocardial revascularization
Time period
Operative deaths
Patients having second revasc. procedure
January, 1970, to July, 1973 July, 1973, to July, 1977
No.
15 33
1%
5
33 3
I
Table VI. Long-term follow-up on patients surviving repeated myocardial revascularization Group 1* (/5 survivors)
I
Group lIt (27 survivors)
I
%
No.
13
87
23
85
2
13
4
15
Status of angina
No.
Angina absent or significantly decreased Angina present
%
'Recurrent angina within the first 2 postoperative months. tRecurrent angina after the second postoperative month.
Secondary procedures performed and postoperative status. It was possible to reconstruct 30 of the 34 (88 percent) occluded grafts in Group I; four (12 percent) grafts were not revised (two because the vessels were not technically satisfactory, one vessel could not be found in dense scar, and one patient with occluded grafts had a cardiac arrest at the induction of anesthesia and could not be resuscitated). In Group 11,34 of the 41 occluded grafts (82 percent) were reconstructed. Seven grafts were not revised (three vessels were too small for grafting, two vessels were diffusely diseased, and two vessels could not be located because of dense scarring).
The internal mammary artery or saphenous vein was used in the majority of patients. A vein patch angioplasty technique was particularly useful in patients with widely patent vein grafts and anastomotic strictures caused by progression of disease, epicardial scar stricture, or technical factors. A total of 13 patients had new nati ve coronary artery lesions bypassed. In Group I only one of the patients (5 percent) was noted to have a new native lesion. In contrast, 12 of the 30 patients in Group II (47 percent) had new native lesions and nearly all were successfully bypassed. The occurrence of postoperative infarction was the same for Groups I and II. Four patients in the study (8 percent) required re-exploration for postoperati ve hemorrhage.
Factors responsible for the development of recurrent angina early (Group I) and late (Group II) after initial revascularization (Table I V). Factors responsible for failure after original revascularization were derived by review of the operative notes, pathology reports, and preoperative angiograms. In most of the patients it was possible to assign one predominant factor responsible for the recurrence of severe angina pectoris. Occlusion by dense pericardial scar. Six patients (33 percent) in the early-onset group and three patients (10 percent) in the late-onset group were found to have exuberant and dense pericardial scar obstructing previously inserted grafts. Incorrect site chosen for distal anastomoses. Placement of original bypass grafts into nonobstructed distal vessels or proximal to stenotic lesions constituted another reason for failure. This error appeared in four patients (22 percent) of the early-onset group and seven patients (23 percent) in the late-onset group. Technical failures of anastomoses. Improper anastomotic technique was a common cause of failure. Commonly, this involved improper placement of sutures at the toe or heel (or both) of the anastomoses, resulting in ultimate occlusion of the graft. Occasionally, the coronary vessel had been opened either obliquely or eccentrically, so that the anastomosis had been difficult and technically awkward to construct. As expected, this factor was noted in six patients (33 percent) of the early-onset group and in only four patients (13 percent) in the late-onset group. Progressive obliteration of the graft. Some saphenous vein grafts were noted to be obstructed by dense myxoid proliferation. This uncommon complication occurred in a total of three patients for both groups. Appearance of new native coronary lesions. Changes
Volume 77 Number 6 June, 1979
in the native coronary circulation were detected by comparing the coronary angiogram before each revascularization procedure. New lesions were noted in 14 patients (47 percent) of the late-onset group and only one patient (6 percent) of the early-onset group. Operative deaths (Table V). Deaths before discharge of the patient from the hospital or within the first postoperative month were considered as operative deaths. A total of six deaths fulfilled these requirements (12.5 percent). When the time period of this study, January, 1970, to August, 1978, was divided into an initial phase-January, 1970, to July, 1973-and a late phase-August, 1973, to July, 1978-a marked change in operative mortality rate was seen. During the first half of this study, five of 15 patients died, a mortality rate of 33 percent. During the last half of the study, only one of 33 patients died, for a mortality rate of 3 percent. Four deaths were due to myocardial infarctionthree occurring in the operating room and one during the second postoperative month. One patient died of a massive pulmonary embolus and one patient died of sepsis from pulmonary complications. Improved methods in anesthetic technique, intraoperative monitoring, and myocardial preservation during coronary bypass grafts are responsible for the significant and gratifying decrease in the operative mortality rate. Late results (Table VI). Follow-up data on all 42 surviving patients were obtained during outpatient follow-up and phone conversations with the referring physician. Status of angina. Thirty-six of the 42 patients (86 percent) were experiencing no angina or rare episodes of angina associated with extreme exertion. When assessed on the basis of the interval between original operation and the reappearance of angina, no appreciable difference was noted. Late deaths. There were two late deaths. One patient died suddenly at home 2'12 years postoperatively of a presumed myocardial infarction. The second patient died of sepsis 7 months following the operation. Discussion From January, 1970, to July, 1978, more than 1,500 patients with severe angina refractory to medical management underwent direct myocardial revascularization at New York University Medical Center. Gratifying symptomatic improvement was noted in 92 percent of the patients with a minimal operative risk. Long-term analysis of our data confirms the impression that coronary artery bypass grafting increases longevity for pa-
Angina following myocardial revascularization
893
tients with multiple coronary obstructions." This observation has been confirmed by others.t' During the same time period 48 patients (3.2 percent) required a second direct revascularizing procedure for the reappearance of incapacitating angina. A careful analysis of these patients has been conducted to answer two fundamental questions: First, what were the specific factors responsible for initial surgical failure? Second, who will benefit from a second procedure and at what risk? Patients who begin having recurrent angina within 2 months of operation (Group I) and those with the later onset of angina (Group II) were noted to differ in several significant ways. Since the time of onset of recurrent pain is the single objective parameter available when patients are initially evaluated, this means a classification is practical and provides valuable prognostic information. Factors predisposing to initial surgical failure. Preoperative factors prior to initial revascularization. Adequate and complete assessment of native coronary circulation is a prerequisite for properly planned revascularization. The coronary vessels should be visualized in multiple radiographic exposures to determine the extent and location of atherosclerotic plaques. A careful study is required of vessels that are totally obstructed during direct proximal injection but that are visualized during retrograde filling. These vessels are frequently of large caliber and can be overlooked easily because they fill with small quantities of dye during angiography. Major vessels that are more than 70 percent obstructed should be bypassed. Wukasch and associates" recently have suggested the importance of revascularizing vessels which are only 50 to 70 percent obstructed at the time of initial operation. This suggestion is based on their analysis of 41 patients requiring repeated myocardial revascularization. During initial revascularization 10 arteries with 50 to 70 percent obstructions were not bypassed. Of these, nine had progressed to significant stenosis within 6 months to 4 years after initial operation, and in five patients they were the only reason for reoperation. This concept requires additional documentation before it can be recommended, but the data reveal a striking progression of disease in a relatively short period of time. Intraoperative factors during initial revascularization. Two intraoperative factors determine to a large extent the success of coronary bypass grafting: surgical techniques and completeness of revascularization. Vein graft failure was attributed to purely technical factors in II patients-six in Group I (33 percent) and four in Group II (13 percent). Distal coronary anastomoses were placed proximal to significantly obstruct-
894
Culliford et al.
ed lesions or in the wrong vessel in 11 patients-four in Group I (22 percent) and seven in Group II (23 percent). These factors underscore the importance of precise surgical technique during coronary anastomosis. Coronary arteriotomy should be done distal to significant obstruction in a region where the vessel is uninvolved with atherosclerosis. This site is apparent when the telltale "blue line," representing the lumen distended with blood, is noted in the coronary artery. Following arteriotomy the vessel is sized and gently probed to assure an unobstructed distal lumen. Anastomosis is carried out with a 6-0 to 7-0 polypropylene suture. Optical magnification and the high-intensity head lamp enhance visualization of the fine details of coronary anatomy. Graft obstruction by neointimal hyperplasia with ultimate closure of the lumen was noted in only three patients. This gratifying finding is attributable to gentle dissection of the saphenous vein, distention of the vessel with cold heparinized blood prior to insertion, and minimal direct handling of the intima and adventitia during preparation of the vein graft and actual anastomosis. The completeness of revascularization has been studied carefully in this series of patients. At the time of initial operation each patient received an average of 2.4 grafts compared to the 2.5 obstructed vessels noted at angiography, for a revascularization rate of 96 percent. As a result of this policy, no patient in this series had recurrent and disabling angina attributable to incomplete initial revascularization. This high revascularization rate is in marked contrast to other reported experiences. Johnson and co-workers" noted that of the 39 patients requiring a second operation, only three had triple bypass grafts initially. This group now frequently bypasses vessels which are 40 to 50 percent obstructed at the time that more severe lesions are being bypassed in an effort to achieve complete revascularization. In a series of 219 cases recently reported from the Cleveland Clinic, incomplete initial revascularization was responsible for 18 percent of the secondary operations for angina." Similar experiences have been reported by Winkle and co-workers ,8 with 21 reoperations. In this series 50 percent of the patients were initially treated with single bypass grafting and 40 percent received double bypass grafting. Nine critical lesions were not bypassed at the original operation, all of which were in the circumflex system. The authors noted that patients recei ving "total surgical correction" of obstructed coronary vessels after initial and secondary procedures have significantly greater improvement
The Journal of Thoracic and Cardiovascular Surgery
in the level of angina than those "not totally corrected. " Postoperative factors predisposing to initial surgical failure. Graft occlusion was attributed to the presence of dense pericardial scar in 33 percent of Group I and 10 percent of Group II patients. These findings were noted in patients who exhibited a significant postpericardiotomy reaction following the initial operation. Typically, this was manifested by fever, malaise, prolonged pleuropericardial pain, and a persistent rub. Significant leukocytosis and an abnormal number of lymphocytes were commonly noted during the immediate postoperative course. These findings and symptoms prompt a thorough investigation to exclude other sources of infection. Once they are excluded, therapy with either acetylsalicylic acid or a short course of steroids is instituted. Similar observations and conclusions have been reported by Oglietti ," Adam, 10 and their colleagues. Significant factors during repeated revascularization predisposing to successful outcomes. In the past, resternotomy itself has been a hazardous and sometimes fatal operation. Following the operative guidelines briefly described, resternotomy was performed in the 48 consecutive patients without surgical misadventure necessitating the premature institution of cardiopulmonary bypass. Re-exploration for bleeding was nevertheless necessary in four patients or 8 percent, which is nearly eight times the re-exploration rate for primary sternotomy. Technical revision of occluded grafts was possible in 93 percent of the cases. This was usually accomplished with a new saphenous vein or internal mammary graft; however, 7 percent of occluded grafts were directly revised by vein patch angioplasty. The greater frequency of new native coronary artery lesions in Group II (47 percent) is not surprising and has been noted in other series. 8. 9. 11 Fortunately, all new lesions in both groups could be technically bypassed. The importance of a complete secondary revascularization is of paramount significance in achieving gratifying long-term results. Operative mortality rate and long-term results. Although the over-all mortality rate for the entire series is 12 percent, only one death has occurred in the last 33 consecutive patients (3 percent). As experience has accumulated with this procedure, simultaneous refinements in anesthetic management, intraoperative monitoring, and improved methods of myocardial preservation have enhanced the safety of a second revascularization.
Volume 77 Number 6
Angina following myocardial revascularization
895
June. 1979
Significant and impressive reduction or ablation of angina was noted in long-term survivors. The over-all reduction or ablation of angina is attributable to an aggressive policy of revascularization in patients with failed grafts and new coronary artery lesions. These results compare favorably with other series in which the reported frequency of significant postoperative improvement has ranged from 47 to 90 percent. 5.8.9. 11-13 From these data, it is apparent that repeated myocardial revascularization can be carried out with a negligible operative risk and offers sustained symptomatic relief in 80 to 90 percent of patients. The time of onset of angina after the initial operation is significant, since it is indicative of the cause of recurrent pain and to a degree predicts the outcome. REFERENCES Isom OW, Spencer FC, Glassman E, Cunningham IN, Teiko P. Reed GE, Boyd AD: Does coronary bypass increase longevity? J THORAC CARDIOVASC SURG 70:28-37, 1978 2 Stiles QR, Lindesmith GG, Tuckes BL, Hughes RK, Meyer BW: Long-term follow-up of patients with coronary artery grafts. Circulation 54:Suppl 3:32, 1976 3 Sheldon WC, Loop FD: Direct myocardial revascularization-1976. Cleve Clin Q 43:97-108, 1976 4 Tecklenberg PL, Alderman EL, Miller DC, Shumway NE, Harrison DC: Changes in survival and symptom relief in a longitudinal study of patients after bypass surgery. Circulation 51, 52:Suppl 1:98, 1975
5 Wukasch DC, Toscano M, Cooley DA, Reul GJ, Sandford FM, Kyger ER, Hallman GL: Reoperation following direct myocardial revascularization. Circulation 56:Suppl 2:3, 1977 6 Johnson WD, Hoffman JF Jr, Flemma RJ, Tector AJ: Secondary surgical procedure for myocardial revascularization J THORAC CARDIOVASC SURG 64:523-529, 1972 7 lrarrazaval MJ, Cosgrove DM, Loop FD, Ennix CL Jr, Groves LK, Taylor PC: Reoperation for myocardial revascularization. J THORAC CARDIOVASC SURG 73: 181188, 1977 8 Winkle RA, Alderman EL, Shumway NE, Harrison DC: Results of reoperation for unsuccessful coronary artery bypass surgery. Circulation 51:Suppl 1:62, 1975 9 Oglietti J, Angelini P, Leachman RD, Cooley DA: Myocardial revascularization. Early and late results after reoperation. J THORAC CARDIOVASC SURG 71:736-740, 1976 10 Adam M, Geisler GF, Lambert CJ, Mitchel BF: Reoperation following clinical failure of aorta-to-coronary artery bypass vein grafts. Ann Thorac Surg 14:272-28 I, 1972 II Thomas CS, Alford WC, Burrus GR, Frist RA, Stoney WS: Results of reoperation for failed aortocoronary bypass grafts. Arch Surg 111:1210-1213, 1976 12 Norwood WI, Cohn LH, Collins 11 Jr: Results of reoperation for recurrent angina pectoris. Ann Thorac Surg 23:9-13, 1977 13 Loop FD, Carabajal NR, Taylor PC, Irarrazaval MJ: Internal mammary artery bypass grafts in reoperative myocardial revascularization. Am J Cardiol 37:890-895, 1976