- Email: [email protected]
Role of myocardial revascularization in postinfarction ventricular septal rupture
Role of Myocardial Revascularization in Postinfarction Ventricular Septal Rupture Rene´ Preˆtre, MD, Qing Ye, MD, Jurg Gru¨nenfelder, MD, Gregor Zund, MD, and Marko I. Turina, MD Clinic for Cardiovascular Surgery, University Hospital Zu¨rich, Zu¨rich, Switzerland
Background. Postinfarction ventricular septal rupture requires urgent closure. The role of systematic coronarography and coronary revascularization needs clarification. Methods. Fifty-four patients who underwent patch closure of postinfarction ventricular septal defect were reviewed. A coronarography had been systematically and myocardial revascularization selectively (when significant coronary artery stenosis existed) performed. Results. No patient died or deteriorated during coronarography. Twenty-six patients showed no coronary artery disease besides the infarct-related artery, and 28 had associated disease. Threatened myocardial territories were revascularized usually with venous grafts (mean
number of distal anastomosis, 2.5). Operative mortality was 19% and 32% (p ⴝ 0.36) and late mortality 43% and 53% (p ⴝ 0.75) in patients without and in patients with associated coronary artery disease, respectively. Survival curve in both group was similar, at least up to 8 years after operation. Conclusions. Myocardial revascularization controlled the added risk of associated coronary artery disease in the postoperative period and in median term. A coronarography should be performed in all patients who can be stabilized hemodynamically and myocardial revascularization performed in case of significant stenosis. (Ann Thorac Surg 2000;69:51–5) © 2000 by The Society of Thoracic Surgeons
R
patients with and of those without revascularization. We assumed that, without revascularization, the postoperative and later course of patients with associated coronary artery disease would have been worse than in patients without associated coronary artery disease.
upture of the septum is a dire complication of myocardial infarction [1– 4]. It appears a few days after myocardial infarction and rapidly produces a severe and progressive state of shock. Early surgical closure of the septal defect is the sole efficient treatment able to reverse the hemodynamic deterioration and frequently appears as a lifesaving procedure [1– 6]. Because the time factor may be critical, some investigators have questioned the wisdom to perform a coronarography during which adequate monitoring and care are hard to provide [5]. A significant number of patients with septal rupture do not present an extensive disease of the coronary arteries beside the occluded infarct artery, and will not draw any benefit from a policy of routine coronarography, whereas other patients, those with additional coronary artery disease, may draw a significant benefit from appropriate revascularization. The risks and benefits of a policy to perform routine coronarography and subsequent targeted myocardial revascularization are not clearly established in patients with septal rupture. Scientifically, the risks and benefit of this policy could only be evaluated if a control group of patients (patients who would not have been sent to coronarography but instead directly to operating room) existed. Such a group does not exist in our praxis. We tried, nonetheless, to bring some light on this problematic by analyzing the morbidity related to the coronarography and by comparing the results of Accepted for publication May 6, 1999. Address reprint requests to Dr Preˆtre, Klinik fu¨r Herzgefa¨sschirurgie, Universita¨tsspital, 100 Ra¨mistrasse, 8091 Zu¨rich, Switzerland; e-mail: [email protected].
© 2000 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
Material and Methods The medical records of 54 consecutive patients admitted in our hospital for a ventricular septum defect after myocardial infarction between 1979 and 1998 were reviewed. Information regarding the characteristics of the coronary arteries were obtained from the angiogram and the operative note reports. Two groups of patients were defined and compared to assess the risks and benefits of performing routine coronarography and selective myocardial revascularization. Group 1 comprises the patients who did not present significant coronary artery disease in addition to the infarct-related artery (patients with one-vessel disease) and group 2 comprises the patients with additional coronary artery disease (patients with two- and threevessel disease). Preoperative clinical data are summarized in Table 1. The diagnosis of ventricular septal rupture was established or confirmed by transthoracic or transesophageal echocardiography, or both, in all patients. Hemodynamic shock, defined as a systolic blood pressure less than 80 mm Hg and reduced peripheral perfusion, occurred in 26 patients (46%) and was treated with proper adjustment of pre- and after-load, use of inotropic support, 0003-4975/00/$20.00 PII S0003-4975(99)00857-7
52
PREˆTRE ET AL REVASCULARIZATION IN SEPTAL RUPTURE
Ann Thorac Surg 2000;69:51–5
Table 1. Preoperative Demographic and Clinical Characteristics of Patients With Postinfarction Septal Rupture Characteristics Men/women Median age (yr) Median delay MI and SR (days) Anteroapical SR Posterobasal SR Hemodynamic shocka Median delay SR and operation (days)
Whole Series (n ⫽ 54)
Group 1 (n ⫽ 26)
Group 2 (n ⫽ 28)
33/21 68 5
14/12 69 4
19/9 66 5
24 30 26 2
14 12 14 2
10 18 12 2
a
Defined as a systolic pressure less than 80 mm Hg and signs of poor peripheral perfusion. MI ⫽ myocardial infarction;
SR ⫽ septal rupture.
and, if these measures were insufficient, with the insertion of an intraaortic balloon pump. If perfusion pressure still remained borderline or the patient remained agitated (agitation was primarily regarded as a sign of reduced cerebral perfusion), tracheal intubation was performed to improve oxygenation and reduce respiratory work, and deep sedation achieved to reduce overall metabolic rate. After appropriate surveillance (once hemodynamic stabilization was obtained), the transfer of the patient to the angiographic suite was organized. At the beginning of the experience, the angiographic examination comprised a cine angiography of the left ventricle and a coronarogram. Later, cine angiography was seldom performed (and definitely avoided in patients with a balloon pump and in those with reduced renal function) to reduce the load of contrast product and the length of the procedure. Upon completion of the angiogram, most of the patients were wheeled back to the intensive care unit before being sent to the next available operating room. A direct transfer to the operating room was organized for the patients who were in borderline hemodynamic condition after insertion of the intraaortic balloon pump. The operation was performed with cardiopulmonary bypass, in moderate hypothermia. The septal defect was usually repaired before performing the coronary artery bypass grafts. The rupture was approached through a left ventriculotomy (except in 2 patients with a chronic rupture and a tricuspid valve insufficiency who were approached through a right atriotomy). Ventricular fibrillation was induced with epicardial electrical stimulation and the left ventricle opened parallel to the septum. In patients with aortic insufficiency (sometimes induced by luxation of the heart to access a posterobasal rupture) and in the 2 patients with a chronic rupture, the ascending aorta was cross-clamped and cardioplegia of the heart obtained with cold hyperkalemic blood. The septum was closed with a patch of pericardium or Dacron sutured on viable myocardium. Closure of the ventriculotomy was performed with care taken to restore or correct the geometry of the left ventricle. Grafting of the
coronary arteries was performed during cardioplegic arrest of the heart, except in 4 recent patients, where grafting was performed on a beating, unloaded heart. Stenosis more than 50% of the vessel diameter on main coronary artery trunks or large branches were considered for grafting. The infarct artery, however, was not routinely bypassed because it frequently ends up in the suture line. A bypass to this artery was performed if a significant side branch (a septal or diagonal branch of the left anterior descending artery or a marginal branch of the right coronary artery) could be perfused. If a stenosis existed on the origin of a large diagonal or marginal branch, a bypass was performed directly to this branch. Thirteen patients received a bypass to the infarct-related artery or to one of its branches. Eight of these patients did not show additional coronary artery disease and are included in group 1 (one-vessel disease). Median time of cardiopulmonary bypass was 88 minutes (range, 40 to 265 minutes) for the entire series, 67 minutes (range, 40 to 265 minutes) for group 1, and 92 minutes (range, 40 to 185 minutes) for group 2. Median aortic cross-clamping time was 40 minutes (range, 26 to 63 minutes) for 8 patients in group 1, and 48 minutes (range, 23 to 87 minutes) for 24 patients in group 2. The operation was performed without cross-clamping of the aorta in 21 patients (17 and 4 patients in groups 1 and 2, respectively). Eight patients in group 1 received one anastomosis to the infarct artery or one of its branch. The 28 patients in group 2 received an average of 2.5 distal anastomosis (range, 1 to 5). Follow-up was obtained in all patients after discharge and was updated in all survivors during the last trimester of 1998 by telephone contact with their physician or cardiologist. Follow-up ranged from 2 months to 18 years (median, 42 months).
Statistical Analysis Physiologic variables (age older than 70 years, cardiogenic shock, and delay between infarction and septal rupture) and anatomic variables (location of the rupture and associated coronary artery disease) were tested for their significance in predicting operative and late mortality. The variables were compared by the Fisher’s exact test and a p value less than 0.05 was considered significant. Survival curves were calculated by the KaplanMeier method.
Results All patients could be stabilized with appropriate medical support: insertion of an intraaortic balloon pump in 26 patients, and further tracheal intubation and deep sedation in 6 patients. They all underwent coronarography without hemodynamic deterioration. Two patients deteriorated 2 and 3 hours after coronarography. Both remained stable during the investigation and during the subsequent 2 hours of surveillance that was observed at the intensive care unit before transfer to the operating room. The first patient showed deterioration during transfer to the operating room and the second one during sternotomy. The first patient required mechanical mas-
PREˆTRE ET AL REVASCULARIZATION IN SEPTAL RUPTURE
Ann Thorac Surg 2000;69:51–5
Table 2. Cause of Postoperative Death Cause of Postoperative Death
Whole Series (n ⫽ 54)
Group 1 (n ⫽ 26)
Group 2 (n ⫽ 28)
9 2 1 1 1 14
3 1
6 1 1 1
a
Myocardial failure Sepsis Left ventricle rupture Postoperative bleeding Sudden death Total a
1 5
9
Includes two intraoperative deaths.
sage. Both patients were swiftly cannulated and supported by cardiopulmonary bypass. However, weaning from cardiopulmonary bypass was impossible because of refractory myocardial failure and both patients died intraoperatively. The rupture of the septum occurred in the anteroapical septum in the 24 patients with an occlusion of the left anterior descending artery and in the posterobasal septum in the 30 patients with an occlusion of the right coronary artery. Twenty-six patients presented with single-vessel disease (the infarct-related artery) and 28 with multiple-vessel disease, among whom 19 patients had two-vessel disease and 9 patients three-vessel disease. Fourteen patients (including 2 patients who died intraoperatively) died during hospitalization (postoperative mortality, 26%), and 19 after hospital discharge. The cause of postoperative death is summarized in Table 2. The postoperative morbidity of the 40 operative survivors was as follows: prolonged intubation (⬎ 7 days), 12 patients; renal failure (hemofiltration), 7; postoperative bleeding, 5; sepsis, 5; transient low cardiac output, 4, neurologic deficit, 1 patient. The cause of late death is shown in Table 3. The predictive value of preoperative factors for postoperative and late death is presented in Table 4. Associated coronary artery disease had a marginally negative impact on postoperative death (p ⫽ 0.36) and no impact on late death (p ⫽ 0.75). A short delay between occurrence of septal rupture and surgical treatment was the sole significant predictor of poor outcome (p ⫽ 0.03 for postoperative death and 0.23 for late death). Preoperative shock showed only a trend for increased postoperative mortality (p ⫽ 0.13) and no influence on late death (p ⫽ 0.52). Survival curves of patients with and
53
Table 4. Predictors of Increased Postoperative and Late Death Rate
Variables Age ⬍ 70 yr ⬎ 70 yr Early experience 1979 –1988 1989 –1998 Preoperative shock No Yes Delay between SR and operation ⬍ 36 h ⬎ 36 h Associated CAD No (group 1) Yes (group 2) Location of SR Anteroapical Posterobasal
Postoperative Mortality Fisher’s Late Fisher’s (%) Test Mortality Test 6/31 (19) 8/23 (35)
0.23
9/25 10/15
7/21 (33) 7/33 (21)
0.36
10/14 9/26
5/29 (17) 9/25 (36)
0.13
10/24 9/16
0.52
12/30 (40) 3/24 (13)
0.03
7/19 12/21
0.23
5/26 (19) 9/28 (32)
0.36
9/21 10/19
0.75
4/24 (21) 10/30 (30)
0.22
8/20 11/20
0.37
0.10
a
a
p value was not calculated because of the difference in the observation periods. CAD ⫽ coronary artery disease;
SR ⫽ septal rupture.
without associated coronary artery disease are depicted in Figure 1. No difference appeared between the two groups, although the patients of group 1 (without associated coronary artery disease) tended to fare better than those of group 2, especially after 8 years. The small number of patients at risk after 8 years reduces the statistical power of the comparison and no longer allows definitive conclusions. It is noteworthy that 2 patients in group 2 died of a myocardial infarction 8 and 10 years after operation.
Table 3. Cause of Late Death Cause of Late Death Myocardial failure Sudden death Myocardial infarction Septal patch endocarditis Neurologic Noncardiac (carcinoma, hepatitis) Total
Whole Series (n ⫽ 40)
Group 1 (n ⫽ 21)
Group 2 (n ⫽ 19)
9 3 2 1 1 3
5 1 2 1 1 1
4 2
2
19
11
8
Fig 1. Survival of patients with and without associated coronary artery disease (CAD).
54
PREˆTRE ET AL REVASCULARIZATION IN SEPTAL RUPTURE
Comment A conspicuous finding of our study is that a significant number of patients (26 of 54 patients) with septal rupture after myocardial infarction did not present significant stenosis on the coronary arteries beside the infarctrelated artery. We performed a coronarography in all our patients and a bypass to the major coronary arteries or side branches when a 50% stenosis existed. The infarct artery, because it frequently becomes entrapped in the suture line of the ventriculotomy, was not routinely bypassed, unless perfusion of one of its proximal branches could be improved. Approximately one-half of our patients received a bypass to a coronary artery supplying a remote myocardial territory. This policy seems to have protected them from subsequent myocardial ischemic events during the postoperative period and during follow-up. Mortality rates were identical between patients with and without associated coronary artery disease. This finding suggests that the added risk of coronary artery disease was controlled by the bypasses performed. The myocardial protection conferred by the bypasses, however, seems to fade with years. Although the number of patients at risk after 8 years is too small to allow meaningful comparison, it seems that the survival curves of the two groups diverge notably from each other. It is further perplexing to note that 2 of the patients who received coronary artery bypass died of myocardial infarction after 8 years. Progression or development of atherosclerosis on the native arteries and grafts may account for these infarcts and for the reduced survival. The rapidity with which the hemodynamic state of a patient with septal rupture deteriorates has led some surgeons to renounce coronarography for fear of wasting precious time and therefore, reducing the chances of survival of their patients [5, 7]. Skillington and associates [5], for instance, warn against a prohibitive mortality and morbidity of the sole coronarography. According to their experience, 4 of their 101 patients with septal rupture died during this examination and a few other patients developed or worsened renal insufficiency after the use of contrast product. The renal insufficiency contributed, in turn, to the later demise of additional patients. Because the diagnosis of septal rupture is now rapidly and reliably established by echocardiography, they strongly advise against systematic coronarography (and therefore, set aside the opportunity to perform targeted bypass grafting), especially in unstable patients. So a grim experience with coronarography has not been shared by other investigators [2– 4, 8 –10]. Like us, David [2], Deville [3], and Cox [11] and their associates, in relatively large series, did not lose patients because of the coronarography. The insertion of an intraaortic balloon pump in every patient requiring inotropic support, and the early intubation and deep sedation of critical patients allowed us to stabilize the hemodynamic of all our patients and perform a coronarography without undue risk. To reduce the renal toxicity due to contrast product, our cardiologists no longer perform a ventriculography and limit the procedure to a few views of the coronary arteries. Still, 2
Ann Thorac Surg 2000;69:51–5
patients in our series deteriorated before surgical correction, and later could not be weaned from cardiopulmonary bypass. The deterioration occurred a few hours after the coronarography and was not linked to this examination. These cases stress the precarious hemodynamic state of these patients and the need for rapid surgical correction. We support the policy of performing a coronarography in every patient that can be adequately stabilized and should renounce it in those patients who remain unstable or require increasing hemodynamic support after insertion of an intraaortic balloon pump. Grafting stenotic coronary arteries seems to have favorably influenced the short- and median-term prognosis of our patients with associated coronary artery disease, because their postoperative and late mortality rates were not worse than those of patients without associated coronary artery disease. This fact seems particularly relevant if one considers the postoperative course and survival of patients with coronary artery disease who underwent a vascular procedure [12, 13]. The mortality and incidence of myocardial infarction during the postoperative period was three times higher in patients with untreated coronary artery disease than in patients without coronary artery disease. The coronary risk was annihilated by concomitant myocardial revascularization [12, 13]. Similarly, the survival of patients with untreated coronary artery disease was worse than the survival of patients without coronary artery disease, with a difference appearing already statistically significant after a few years [12, 13]. Again, concomitant myocardial revascularization at the time of the peripheral procedure improved survival to values identical to those achieved by patients without coronary artery disease [14]. Although any extrapolation remains delicate, it intuitively seems probable that the coronary risk might be even greater in our patients who were in shock, underwent a major cardiac operation, and frequently required significant inotropic and vasopressive support.
Limitation of the Study Because of the characteristics of this study (retrospective analysis of patients who all underwent systemic coronarography), it is impossible to scientifically demonstrate the benefit of concomitant myocardial revascularization in patients with associated coronary artery disease. Only a comparison with a control group (ie, patients who would not undergo a coronarography and therefore, a targeted revascularization) would allow such a demonstration. The demonstration is unlikely to ever be performed, because as demonstrated here, the majority of patients with septal rupture can safely undergo a coronarography. It would then become unethical to refrain from performing bypass to myocardial territories threatened by a coronary stenosis. Our study, however, allows us to state that coronarography can be performed safely in the majority of patients, and that revascularization controls the coronary risk at least up to the median term. In conclusion, revascularization of the myocardium at the time of septal closure after myocardial infarction allows the control of the added coronary risk and mor-
PREˆTRE ET AL REVASCULARIZATION IN SEPTAL RUPTURE
Ann Thorac Surg 2000;69:51–5
tality during the postoperative and median-term periods. The long-term prognosis is, however, less certain, probably due to the progression or development of atherosclerotic disease on the native coronary arteries or bypass grafts. Because inherent mortality and morbidity are not prohibitive, and because it allows a targeted revascularization, a coronarography seems indicated in every patient that can be hemodynamically stabilized.
References 1. Cox FF, Morshuis WJ, Plokker HWM, et al. Early mortality after surgical repair of postinfarction ventricular septal rupture: importance of rupture location. Ann Thorac Surg 1996; 61:1752– 8. 2. David TE, Dale L, Sun Z. Postinfarction ventricular septal rupture: repair by endocardial patch with infarct exclusion. J Thorac Cardiovasc Surg 1995;110:1315–22. 3. Deville C, Fontan F, Chevalier JM, Madonna F, Ebner A, Besse P. Surgery of post-infarction ventricular septal defect: risk factors for hospital death and long-term results. Eur J Cardiothorac Surg 1991;5:167–75. 4. Von Segesser LK, Siebenmann R, Schneider K, Jenni R, Gallino A, Turina MI. Post-infarction ventricular septal defect: surgical strategies and results. Cardiovasc Thorac Surg 1989;37:72–5. 5. Skillington PD, Davies RH, Luff AJ, et al. Surgical treatment for infarct-related ventricular septal defect. J Thorac Cardiovasc Surg 1990;99:798 – 808. 6. Topaz O, Taylor AL. Interventricular septal rupture compli-
7. 8. 9. 10. 11.
12.
13.
14.
55
cating acute myocardial infarction: from pathophysiologic features to the role of invasive and noninvasive diagnostic modalities in current management. Am J Med 1992;93:683– 8. Muehrcke DD, Daggett WM, Buckley M, et al. Postinfarction ventricular septal defect repair: effect of coronary artery bypass grafting. Ann Thorac Surg 1992;54:876– 83. Daggett WM. Surgical technique for early repair of posterior ventricular septal rupture. J Thorac Cardiovasc Surg 1982;84: 306–12. Radford MJ, Johnson RA, Daggett WM, et al. Ventricular septal rupture: a review of clinical and physiological features and an analysis of survival. Circulation 1981;64:545–53. Loisance DY, Lordez JM, Deleuze PH, Dubois-Rande JL, Lellouche D, Cachera JP. Acute postinfarction septal rupture: long-term results. Ann Thorac Surg 1991;52:474– 8. Cox FF, Plokker HWM, Morshuis WJ, Kelder JC, Vermeulen FE. Importance of coronary revascularisation for late survival after post-infarction ventricular septal rupture. Eur Heart J 1996;17:1841–5. Roger VL, Ballard DJ, Hallet JW, et al. Influence of coronary artery disease on morbidity and mortality after abdominal aortic aneurysmectomy: a population-based study, 1971– 1987. J Am Coll Cardiol 1989;14:1245–52. Farkouh ME, Rihal CS, Gersh BJ, et al. Influence of coronary artery disease on morbidity and mortality after lower extremity revascularisation: a population-based study in Olmsted County, Minnesota (1970 –1987). J Am Coll Cardiol 1994; 24:1290– 6. Hertzer NR, Young JR, Kramer JR, et al. Routine coronary angiography prior to elective myocardial revascularization in patients with peripheral vascular disease. Arch Surg 1979; 114:1336– 44.
Background. Postinfarction ventricular septal rupture requires urgent closure. The role of systematic coronarography and coronary revascularization needs clarification. Methods. Fifty-four patients who underwent patch closure of postinfarction ventricular septal defect were reviewed. A coronarography had been systematically and myocardial revascularization selectively (when significant coronary artery stenosis existed) performed. Results. No patient died or deteriorated during coronarography. Twenty-six patients showed no coronary artery disease besides the infarct-related artery, and 28 had associated disease. Threatened myocardial territories were revascularized usually with venous grafts (mean
number of distal anastomosis, 2.5). Operative mortality was 19% and 32% (p ⴝ 0.36) and late mortality 43% and 53% (p ⴝ 0.75) in patients without and in patients with associated coronary artery disease, respectively. Survival curve in both group was similar, at least up to 8 years after operation. Conclusions. Myocardial revascularization controlled the added risk of associated coronary artery disease in the postoperative period and in median term. A coronarography should be performed in all patients who can be stabilized hemodynamically and myocardial revascularization performed in case of significant stenosis. (Ann Thorac Surg 2000;69:51–5) © 2000 by The Society of Thoracic Surgeons
R
patients with and of those without revascularization. We assumed that, without revascularization, the postoperative and later course of patients with associated coronary artery disease would have been worse than in patients without associated coronary artery disease.
upture of the septum is a dire complication of myocardial infarction [1– 4]. It appears a few days after myocardial infarction and rapidly produces a severe and progressive state of shock. Early surgical closure of the septal defect is the sole efficient treatment able to reverse the hemodynamic deterioration and frequently appears as a lifesaving procedure [1– 6]. Because the time factor may be critical, some investigators have questioned the wisdom to perform a coronarography during which adequate monitoring and care are hard to provide [5]. A significant number of patients with septal rupture do not present an extensive disease of the coronary arteries beside the occluded infarct artery, and will not draw any benefit from a policy of routine coronarography, whereas other patients, those with additional coronary artery disease, may draw a significant benefit from appropriate revascularization. The risks and benefits of a policy to perform routine coronarography and subsequent targeted myocardial revascularization are not clearly established in patients with septal rupture. Scientifically, the risks and benefit of this policy could only be evaluated if a control group of patients (patients who would not have been sent to coronarography but instead directly to operating room) existed. Such a group does not exist in our praxis. We tried, nonetheless, to bring some light on this problematic by analyzing the morbidity related to the coronarography and by comparing the results of Accepted for publication May 6, 1999. Address reprint requests to Dr Preˆtre, Klinik fu¨r Herzgefa¨sschirurgie, Universita¨tsspital, 100 Ra¨mistrasse, 8091 Zu¨rich, Switzerland; e-mail: [email protected].
© 2000 by The Society of Thoracic Surgeons Published by Elsevier Science Inc
Material and Methods The medical records of 54 consecutive patients admitted in our hospital for a ventricular septum defect after myocardial infarction between 1979 and 1998 were reviewed. Information regarding the characteristics of the coronary arteries were obtained from the angiogram and the operative note reports. Two groups of patients were defined and compared to assess the risks and benefits of performing routine coronarography and selective myocardial revascularization. Group 1 comprises the patients who did not present significant coronary artery disease in addition to the infarct-related artery (patients with one-vessel disease) and group 2 comprises the patients with additional coronary artery disease (patients with two- and threevessel disease). Preoperative clinical data are summarized in Table 1. The diagnosis of ventricular septal rupture was established or confirmed by transthoracic or transesophageal echocardiography, or both, in all patients. Hemodynamic shock, defined as a systolic blood pressure less than 80 mm Hg and reduced peripheral perfusion, occurred in 26 patients (46%) and was treated with proper adjustment of pre- and after-load, use of inotropic support, 0003-4975/00/$20.00 PII S0003-4975(99)00857-7
52
PREˆTRE ET AL REVASCULARIZATION IN SEPTAL RUPTURE
Ann Thorac Surg 2000;69:51–5
Table 1. Preoperative Demographic and Clinical Characteristics of Patients With Postinfarction Septal Rupture Characteristics Men/women Median age (yr) Median delay MI and SR (days) Anteroapical SR Posterobasal SR Hemodynamic shocka Median delay SR and operation (days)
Whole Series (n ⫽ 54)
Group 1 (n ⫽ 26)
Group 2 (n ⫽ 28)
33/21 68 5
14/12 69 4
19/9 66 5
24 30 26 2
14 12 14 2
10 18 12 2
a
Defined as a systolic pressure less than 80 mm Hg and signs of poor peripheral perfusion. MI ⫽ myocardial infarction;
SR ⫽ septal rupture.
and, if these measures were insufficient, with the insertion of an intraaortic balloon pump. If perfusion pressure still remained borderline or the patient remained agitated (agitation was primarily regarded as a sign of reduced cerebral perfusion), tracheal intubation was performed to improve oxygenation and reduce respiratory work, and deep sedation achieved to reduce overall metabolic rate. After appropriate surveillance (once hemodynamic stabilization was obtained), the transfer of the patient to the angiographic suite was organized. At the beginning of the experience, the angiographic examination comprised a cine angiography of the left ventricle and a coronarogram. Later, cine angiography was seldom performed (and definitely avoided in patients with a balloon pump and in those with reduced renal function) to reduce the load of contrast product and the length of the procedure. Upon completion of the angiogram, most of the patients were wheeled back to the intensive care unit before being sent to the next available operating room. A direct transfer to the operating room was organized for the patients who were in borderline hemodynamic condition after insertion of the intraaortic balloon pump. The operation was performed with cardiopulmonary bypass, in moderate hypothermia. The septal defect was usually repaired before performing the coronary artery bypass grafts. The rupture was approached through a left ventriculotomy (except in 2 patients with a chronic rupture and a tricuspid valve insufficiency who were approached through a right atriotomy). Ventricular fibrillation was induced with epicardial electrical stimulation and the left ventricle opened parallel to the septum. In patients with aortic insufficiency (sometimes induced by luxation of the heart to access a posterobasal rupture) and in the 2 patients with a chronic rupture, the ascending aorta was cross-clamped and cardioplegia of the heart obtained with cold hyperkalemic blood. The septum was closed with a patch of pericardium or Dacron sutured on viable myocardium. Closure of the ventriculotomy was performed with care taken to restore or correct the geometry of the left ventricle. Grafting of the
coronary arteries was performed during cardioplegic arrest of the heart, except in 4 recent patients, where grafting was performed on a beating, unloaded heart. Stenosis more than 50% of the vessel diameter on main coronary artery trunks or large branches were considered for grafting. The infarct artery, however, was not routinely bypassed because it frequently ends up in the suture line. A bypass to this artery was performed if a significant side branch (a septal or diagonal branch of the left anterior descending artery or a marginal branch of the right coronary artery) could be perfused. If a stenosis existed on the origin of a large diagonal or marginal branch, a bypass was performed directly to this branch. Thirteen patients received a bypass to the infarct-related artery or to one of its branches. Eight of these patients did not show additional coronary artery disease and are included in group 1 (one-vessel disease). Median time of cardiopulmonary bypass was 88 minutes (range, 40 to 265 minutes) for the entire series, 67 minutes (range, 40 to 265 minutes) for group 1, and 92 minutes (range, 40 to 185 minutes) for group 2. Median aortic cross-clamping time was 40 minutes (range, 26 to 63 minutes) for 8 patients in group 1, and 48 minutes (range, 23 to 87 minutes) for 24 patients in group 2. The operation was performed without cross-clamping of the aorta in 21 patients (17 and 4 patients in groups 1 and 2, respectively). Eight patients in group 1 received one anastomosis to the infarct artery or one of its branch. The 28 patients in group 2 received an average of 2.5 distal anastomosis (range, 1 to 5). Follow-up was obtained in all patients after discharge and was updated in all survivors during the last trimester of 1998 by telephone contact with their physician or cardiologist. Follow-up ranged from 2 months to 18 years (median, 42 months).
Statistical Analysis Physiologic variables (age older than 70 years, cardiogenic shock, and delay between infarction and septal rupture) and anatomic variables (location of the rupture and associated coronary artery disease) were tested for their significance in predicting operative and late mortality. The variables were compared by the Fisher’s exact test and a p value less than 0.05 was considered significant. Survival curves were calculated by the KaplanMeier method.
Results All patients could be stabilized with appropriate medical support: insertion of an intraaortic balloon pump in 26 patients, and further tracheal intubation and deep sedation in 6 patients. They all underwent coronarography without hemodynamic deterioration. Two patients deteriorated 2 and 3 hours after coronarography. Both remained stable during the investigation and during the subsequent 2 hours of surveillance that was observed at the intensive care unit before transfer to the operating room. The first patient showed deterioration during transfer to the operating room and the second one during sternotomy. The first patient required mechanical mas-
PREˆTRE ET AL REVASCULARIZATION IN SEPTAL RUPTURE
Ann Thorac Surg 2000;69:51–5
Table 2. Cause of Postoperative Death Cause of Postoperative Death
Whole Series (n ⫽ 54)
Group 1 (n ⫽ 26)
Group 2 (n ⫽ 28)
9 2 1 1 1 14
3 1
6 1 1 1
a
Myocardial failure Sepsis Left ventricle rupture Postoperative bleeding Sudden death Total a
1 5
9
Includes two intraoperative deaths.
sage. Both patients were swiftly cannulated and supported by cardiopulmonary bypass. However, weaning from cardiopulmonary bypass was impossible because of refractory myocardial failure and both patients died intraoperatively. The rupture of the septum occurred in the anteroapical septum in the 24 patients with an occlusion of the left anterior descending artery and in the posterobasal septum in the 30 patients with an occlusion of the right coronary artery. Twenty-six patients presented with single-vessel disease (the infarct-related artery) and 28 with multiple-vessel disease, among whom 19 patients had two-vessel disease and 9 patients three-vessel disease. Fourteen patients (including 2 patients who died intraoperatively) died during hospitalization (postoperative mortality, 26%), and 19 after hospital discharge. The cause of postoperative death is summarized in Table 2. The postoperative morbidity of the 40 operative survivors was as follows: prolonged intubation (⬎ 7 days), 12 patients; renal failure (hemofiltration), 7; postoperative bleeding, 5; sepsis, 5; transient low cardiac output, 4, neurologic deficit, 1 patient. The cause of late death is shown in Table 3. The predictive value of preoperative factors for postoperative and late death is presented in Table 4. Associated coronary artery disease had a marginally negative impact on postoperative death (p ⫽ 0.36) and no impact on late death (p ⫽ 0.75). A short delay between occurrence of septal rupture and surgical treatment was the sole significant predictor of poor outcome (p ⫽ 0.03 for postoperative death and 0.23 for late death). Preoperative shock showed only a trend for increased postoperative mortality (p ⫽ 0.13) and no influence on late death (p ⫽ 0.52). Survival curves of patients with and
53
Table 4. Predictors of Increased Postoperative and Late Death Rate
Variables Age ⬍ 70 yr ⬎ 70 yr Early experience 1979 –1988 1989 –1998 Preoperative shock No Yes Delay between SR and operation ⬍ 36 h ⬎ 36 h Associated CAD No (group 1) Yes (group 2) Location of SR Anteroapical Posterobasal
Postoperative Mortality Fisher’s Late Fisher’s (%) Test Mortality Test 6/31 (19) 8/23 (35)
0.23
9/25 10/15
7/21 (33) 7/33 (21)
0.36
10/14 9/26
5/29 (17) 9/25 (36)
0.13
10/24 9/16
0.52
12/30 (40) 3/24 (13)
0.03
7/19 12/21
0.23
5/26 (19) 9/28 (32)
0.36
9/21 10/19
0.75
4/24 (21) 10/30 (30)
0.22
8/20 11/20
0.37
0.10
a
a
p value was not calculated because of the difference in the observation periods. CAD ⫽ coronary artery disease;
SR ⫽ septal rupture.
without associated coronary artery disease are depicted in Figure 1. No difference appeared between the two groups, although the patients of group 1 (without associated coronary artery disease) tended to fare better than those of group 2, especially after 8 years. The small number of patients at risk after 8 years reduces the statistical power of the comparison and no longer allows definitive conclusions. It is noteworthy that 2 patients in group 2 died of a myocardial infarction 8 and 10 years after operation.
Table 3. Cause of Late Death Cause of Late Death Myocardial failure Sudden death Myocardial infarction Septal patch endocarditis Neurologic Noncardiac (carcinoma, hepatitis) Total
Whole Series (n ⫽ 40)
Group 1 (n ⫽ 21)
Group 2 (n ⫽ 19)
9 3 2 1 1 3
5 1 2 1 1 1
4 2
2
19
11
8
Fig 1. Survival of patients with and without associated coronary artery disease (CAD).
54
PREˆTRE ET AL REVASCULARIZATION IN SEPTAL RUPTURE
Comment A conspicuous finding of our study is that a significant number of patients (26 of 54 patients) with septal rupture after myocardial infarction did not present significant stenosis on the coronary arteries beside the infarctrelated artery. We performed a coronarography in all our patients and a bypass to the major coronary arteries or side branches when a 50% stenosis existed. The infarct artery, because it frequently becomes entrapped in the suture line of the ventriculotomy, was not routinely bypassed, unless perfusion of one of its proximal branches could be improved. Approximately one-half of our patients received a bypass to a coronary artery supplying a remote myocardial territory. This policy seems to have protected them from subsequent myocardial ischemic events during the postoperative period and during follow-up. Mortality rates were identical between patients with and without associated coronary artery disease. This finding suggests that the added risk of coronary artery disease was controlled by the bypasses performed. The myocardial protection conferred by the bypasses, however, seems to fade with years. Although the number of patients at risk after 8 years is too small to allow meaningful comparison, it seems that the survival curves of the two groups diverge notably from each other. It is further perplexing to note that 2 of the patients who received coronary artery bypass died of myocardial infarction after 8 years. Progression or development of atherosclerosis on the native arteries and grafts may account for these infarcts and for the reduced survival. The rapidity with which the hemodynamic state of a patient with septal rupture deteriorates has led some surgeons to renounce coronarography for fear of wasting precious time and therefore, reducing the chances of survival of their patients [5, 7]. Skillington and associates [5], for instance, warn against a prohibitive mortality and morbidity of the sole coronarography. According to their experience, 4 of their 101 patients with septal rupture died during this examination and a few other patients developed or worsened renal insufficiency after the use of contrast product. The renal insufficiency contributed, in turn, to the later demise of additional patients. Because the diagnosis of septal rupture is now rapidly and reliably established by echocardiography, they strongly advise against systematic coronarography (and therefore, set aside the opportunity to perform targeted bypass grafting), especially in unstable patients. So a grim experience with coronarography has not been shared by other investigators [2– 4, 8 –10]. Like us, David [2], Deville [3], and Cox [11] and their associates, in relatively large series, did not lose patients because of the coronarography. The insertion of an intraaortic balloon pump in every patient requiring inotropic support, and the early intubation and deep sedation of critical patients allowed us to stabilize the hemodynamic of all our patients and perform a coronarography without undue risk. To reduce the renal toxicity due to contrast product, our cardiologists no longer perform a ventriculography and limit the procedure to a few views of the coronary arteries. Still, 2
Ann Thorac Surg 2000;69:51–5
patients in our series deteriorated before surgical correction, and later could not be weaned from cardiopulmonary bypass. The deterioration occurred a few hours after the coronarography and was not linked to this examination. These cases stress the precarious hemodynamic state of these patients and the need for rapid surgical correction. We support the policy of performing a coronarography in every patient that can be adequately stabilized and should renounce it in those patients who remain unstable or require increasing hemodynamic support after insertion of an intraaortic balloon pump. Grafting stenotic coronary arteries seems to have favorably influenced the short- and median-term prognosis of our patients with associated coronary artery disease, because their postoperative and late mortality rates were not worse than those of patients without associated coronary artery disease. This fact seems particularly relevant if one considers the postoperative course and survival of patients with coronary artery disease who underwent a vascular procedure [12, 13]. The mortality and incidence of myocardial infarction during the postoperative period was three times higher in patients with untreated coronary artery disease than in patients without coronary artery disease. The coronary risk was annihilated by concomitant myocardial revascularization [12, 13]. Similarly, the survival of patients with untreated coronary artery disease was worse than the survival of patients without coronary artery disease, with a difference appearing already statistically significant after a few years [12, 13]. Again, concomitant myocardial revascularization at the time of the peripheral procedure improved survival to values identical to those achieved by patients without coronary artery disease [14]. Although any extrapolation remains delicate, it intuitively seems probable that the coronary risk might be even greater in our patients who were in shock, underwent a major cardiac operation, and frequently required significant inotropic and vasopressive support.
Limitation of the Study Because of the characteristics of this study (retrospective analysis of patients who all underwent systemic coronarography), it is impossible to scientifically demonstrate the benefit of concomitant myocardial revascularization in patients with associated coronary artery disease. Only a comparison with a control group (ie, patients who would not undergo a coronarography and therefore, a targeted revascularization) would allow such a demonstration. The demonstration is unlikely to ever be performed, because as demonstrated here, the majority of patients with septal rupture can safely undergo a coronarography. It would then become unethical to refrain from performing bypass to myocardial territories threatened by a coronary stenosis. Our study, however, allows us to state that coronarography can be performed safely in the majority of patients, and that revascularization controls the coronary risk at least up to the median term. In conclusion, revascularization of the myocardium at the time of septal closure after myocardial infarction allows the control of the added coronary risk and mor-
PREˆTRE ET AL REVASCULARIZATION IN SEPTAL RUPTURE
Ann Thorac Surg 2000;69:51–5
tality during the postoperative and median-term periods. The long-term prognosis is, however, less certain, probably due to the progression or development of atherosclerotic disease on the native coronary arteries or bypass grafts. Because inherent mortality and morbidity are not prohibitive, and because it allows a targeted revascularization, a coronarography seems indicated in every patient that can be hemodynamically stabilized.
References 1. Cox FF, Morshuis WJ, Plokker HWM, et al. Early mortality after surgical repair of postinfarction ventricular septal rupture: importance of rupture location. Ann Thorac Surg 1996; 61:1752– 8. 2. David TE, Dale L, Sun Z. Postinfarction ventricular septal rupture: repair by endocardial patch with infarct exclusion. J Thorac Cardiovasc Surg 1995;110:1315–22. 3. Deville C, Fontan F, Chevalier JM, Madonna F, Ebner A, Besse P. Surgery of post-infarction ventricular septal defect: risk factors for hospital death and long-term results. Eur J Cardiothorac Surg 1991;5:167–75. 4. Von Segesser LK, Siebenmann R, Schneider K, Jenni R, Gallino A, Turina MI. Post-infarction ventricular septal defect: surgical strategies and results. Cardiovasc Thorac Surg 1989;37:72–5. 5. Skillington PD, Davies RH, Luff AJ, et al. Surgical treatment for infarct-related ventricular septal defect. J Thorac Cardiovasc Surg 1990;99:798 – 808. 6. Topaz O, Taylor AL. Interventricular septal rupture compli-
7. 8. 9. 10. 11.
12.
13.
14.
55
cating acute myocardial infarction: from pathophysiologic features to the role of invasive and noninvasive diagnostic modalities in current management. Am J Med 1992;93:683– 8. Muehrcke DD, Daggett WM, Buckley M, et al. Postinfarction ventricular septal defect repair: effect of coronary artery bypass grafting. Ann Thorac Surg 1992;54:876– 83. Daggett WM. Surgical technique for early repair of posterior ventricular septal rupture. J Thorac Cardiovasc Surg 1982;84: 306–12. Radford MJ, Johnson RA, Daggett WM, et al. Ventricular septal rupture: a review of clinical and physiological features and an analysis of survival. Circulation 1981;64:545–53. Loisance DY, Lordez JM, Deleuze PH, Dubois-Rande JL, Lellouche D, Cachera JP. Acute postinfarction septal rupture: long-term results. Ann Thorac Surg 1991;52:474– 8. Cox FF, Plokker HWM, Morshuis WJ, Kelder JC, Vermeulen FE. Importance of coronary revascularisation for late survival after post-infarction ventricular septal rupture. Eur Heart J 1996;17:1841–5. Roger VL, Ballard DJ, Hallet JW, et al. Influence of coronary artery disease on morbidity and mortality after abdominal aortic aneurysmectomy: a population-based study, 1971– 1987. J Am Coll Cardiol 1989;14:1245–52. Farkouh ME, Rihal CS, Gersh BJ, et al. Influence of coronary artery disease on morbidity and mortality after lower extremity revascularisation: a population-based study in Olmsted County, Minnesota (1970 –1987). J Am Coll Cardiol 1994; 24:1290– 6. Hertzer NR, Young JR, Kramer JR, et al. Routine coronary angiography prior to elective myocardial revascularization in patients with peripheral vascular disease. Arch Surg 1979; 114:1336– 44.