Evolving Indications for Preoperative Intraaortic Balloon Pump Assistance

Evolving Indications for Preoperative Intraaortic Balloon Pump Assistance John Gunstensen, M.D., Bernard S. Goldman, M.D., Hugh E. Scully, M.D., Victor F. Huckell, M.D., a n d Allan G. Adelman, M.D.

Alterations in the myocardial oxygen supply/ demand equation are intermittently present in patients with acute coronary insufficiency. Subendocardial necrosis is the single most common pathological finding at postmortem examination in these patients [37]. Subendocardial perfusion has been correlated with the endocardial viability ratio (EVR) [3], which is increased by IABP [2, 13, 291. Left main coronary artery stenosis is attended by a high medical and surgical mortality [8, 381. Critical narrowing of the coronary circulation does not permit wide variations in either left ventricular afterload or coronary perfusion pressure; the induction of anesthesia is particularly hazardous, but it is made less so by IABP. Left ventricular dysfunction (defined as an ejection fraction of < 0.40) has been associated with a high operative and late mortality after aortocoronary bypass (ACB) or valve replacement [18], Pulsatile flow during the operation results in better retention of organ function than the nonpulsatile flow of conventional bypass [28]. The intraaortic balloon is one The use of intraaortic balloon pumping (IABP) method of introducing pulsatility into a for the management of cardiogenic shock is an standard cardiopulmonary bypass circuit [271. established procedure [2, 5, 261. Indications for The purpose of this paper is to evaluate and its use preoperatively and electively, however, validate these concepts by a review of our exare in an evolving and controversial state. perience with IABP in 150 patients over a twoPreoperative IABP may be of particular value in year period from October, 1973, to November, three groups of patients: those with the clinical 1975. During this period 550 patients with corosyndrome of acute coronary insufficiency, those nary artery disease were operated upon and 405 with critical left main coronary artery stenosis patients underwent various types of valve re(or its equivalent), and those with impaired left placement. ventricular function. ABSTRACT Over a two-year period about 1,000 operations were performed with cardiopulmonary bypass. Intraaortic balloon pump assistance (IABP) was employed on 150 occasions, and a review of these has permitted clarification of the indications for its use. Sixty patients had IABP for cardiogenic shock either after infarction or after cardiotomy, and 37 (62%) survived. Preoperative IABP in 90 high-risk patients resulted in survival for 79 (88%). The indications for prophylactic IABP included: (1) relief of severe pain, which occurred in 42 patients with acute coronary insufficiency, (2) improvement in the coronary perfusion pressure, which was accomplished in 20 patients with significant left main coronary artery occlusion or its equivalent, and (3) protection of left ventricular function, which was carried out in 28 patients with an LV ejection fraction of less than 0.40. The significance of the preoperative endocardial viability ratio (EVR) in relation to prophylactic IABP was also assessed: an EVR below 0.70 appears to be an indication for preoperative IABP.

Clinical Material and Methods From the Departments of Surgery and Medicine, University of Toronto Faculty of Medicine, and the Division of Cardiovascular Surgery, Toronto General Hospital, Toronto, Ont, Canada. Presented at the Twelfth Annual Meeting of The Society of Thoracic Surgeons, Washington, DC, Jan 26-28, 1976. Address reprint requests to Dr. Goldman, Toronto General Hospital, 101 College St, Toronto, Ont, Canada M5G 1L7.

535

Our initial indications for diastolic counterpulsation were divided into two broad groups, clinical and angiographic (Table 1).The clinical indications were: (1) cardiogenic shock following myocardial infarction or cardiopulmonary bypass, and (2) a syndrome of recurrent acute coronary insufficiency. Cardiogenic shock was

536 The Annals of Thoracic Surgery Vol 22 No 6 December 1976

Table 1 . Clinical and Angiographic lndications for lntraaortic Balloon Pump Assistance at Toronto General Hospital Clinical 1. Casdiogenic shock

infarction

(=After

After cardiotomy 2. Acute coronary insufficiency-all patients

Angiographic 1. LV dysfunction (EF < 0.40)


Valve replacement

2. Left main coronary artery stenosis or equivalent LV = left ventricle; EF = ejection fraction; CABG = coronary artery bypass grafting.

defined by standard MIRV clinical criteria, and acute coronary insufficiency was assessed by the criteria adopted by Gazes and associates [lo]. Angiographic indications for IABP were: (1)left ventricular dysfunction (defined as an ejection fraction of < 0.40), and (2) left main coronary artery stenosis greater than 75% (or the anatomical equivalent). Diastolic counterpulsationt was initiated under local anesthesia in all patients with cardiogenic shock after myocardial infarction and in most patients with acute coronary insufficiency or left main coronary artery stenosis. In the remainder the balloon catheter was inserted after induction of general anesthesia but before sternotomy. In patients in whom counterpulsation had been initiated prior to cardiac catheterization, the opposite femoral artery was used for the passage of catheters appropriate to the investigation. Counterpulsation was temporarily discontinued for a few seconds until the catheter had been advanced beyond the proximal end of the balloon. Appropriate measurements were made of the cardiac output (Fick method) and the intracardiac pressures. Coronary angiograms were undertaken with the Judkins technique. The left ventricular ejection fraction was calculated from a single right anterior oblique projection. Pressures were then recorded during withdrawal of the catheter from the left ventricle to allow calculation of the endocardia1 viability ratio. ‘MIRU

=

Cardiopulmonary bypass with hemodilution and moderate hypothermia to 30°C was used during each operation. Periods of anoxia were limited to 20 minutes each. Ventricular fibrillation with left ventricular venting was performed. Coronary perfusion was used during aortic valve replacement.

Results The overall results of IABP are shown in Table 2. Cardiogenic shock was associated with a crude mortality of 38% (23 patients). Prophylactic preoperative counterpulsation resulted in 6 hospital deaths (7.5% mortality) in patients with coronary artery disease and 5 deaths (50%) in patients with valve disease. Cardiogenic Shock The results for patients with cardiogenic shock are shown in Table 3. CARDIOGENIC SHOCK AFTER MYOCARDIAL I N FARCTION.

Six patients with cardiogenic shock

Table 2. Overall Results of lntraaortic Balloon Pump Assistance Total No.

Deaths

Indications for IABP

of Patients

No.

O/O

Cardiogenic shock

60

23

38.0

80

6

7.5

10

5

50.0

Prophylaxis

Coronary

artery disease Valve disease

myocardial infarction research unit.

tAvco, Hoffman-La Roche, Inc, Cranbury, NJ.

IABP = intraaortic balloon pump assistance.

537 Gunstensen et al: Preoperative Intraaortic Balloon Pump Assist

Table 3 . Results of Intraaortic Balloon Pump Assistance for Cardiogenic Shock Cause of Low Cardiac Output After infarction Pure LV failure No operation Operation With arrhythmia (VTIVF) or mech. defect (MVR, VSD) After cardiotomy Coronary artery disease Valve disease

No. of Patients

Survival

3 3 16

0 2 (67%) 11 (69%)

26 12

20 (77%) 4 (33'/o)

LV = left ventricular; VTlVF = ventricular tachycardia, ventricular fibrillation; MVR = mitral valve regurgitation; VSD = ventricular septal defect.

after myocardial infarction had pure left ventricular power failure without a mechanical defect or arrhythmia. Three of these had inoperable coronary artery disease and global left ventricular dysfunction, and although 2 of them were successfully weaned from IABP, all 3 died in the hospital. Immediate operation permitted survival of 2 of the other 3 patients in whom a pure shock state was present. The overall survival rate in this small group of patients with pure left ventricular power failure was 33%. In 16 patients in whom recurrent ventricular arrhythmia or a mechanical defect such as acute mitral regurgitation or acute ventricular septal defect was present, the operative results were improved, with 69% survival (11patients). CARDIOGENIC

SHOCK

AFTER

shock after valve replacement, 4 survived (33%). The best results were achieved when the patient was transferred to IABP from bypass: the 4 survivors were among the 7 patients so treated. There were no survivors in the group of 5 patients in whom IABP was started in the recovery room. Six of the 8 patients who died had normal coronary arteries.

Acute Coronary lnsufficiency Forty-two patients with recurrent acute coronary insufficiency were treated (Tables 4, 5); they represent about 8% of all patients with coronary artery disease managed by us during this period. Eleven of these patients (Table 4) had IABP prior to angiography; in each instance ischemic pain ceased. Subsequent analysis showed that none of these patients had normal coronary arteries. Six had inoperable coronary artery disease, and 1of them died one week after successful weaning from IABP. The 5 patients who had operable disease underwent successful aortocoronary bypass grafting. Two had left ventricular dysfunction (defined as an ejection fraction < 0.4), and 1of them died suddenly on the day of discharge due to occlusion of a left anterior descending coronary artery graft. Table 4. Significant Angiographic, Hemodynarnic, and Biochemical Data in 11 Patients with Acute Coronary Insufficiency W h o Had In traaortic Balloon Pump Assistance before and after Preoperative Angiography

CARDIOTOMY.

Cardiogenic shock occurring after cardiopulmonary bypass was managed by IABP in 26 patients after coronary artery bypass grafting for stable angina. These 26 patients represent 5% of the patients operated on for stable angina from October, 1973, to November, 1975. Twenty of them 77%) survived to leave the hospital. In 22 of these patients, IABP was initiated in the operating room, where they were weaned directly from cardiopulmonary bypass to diastolic augmentation; the 6 patients who died were in this group. In the remaining 4 patients IABP was started in the recovery room 3 to 24 hours after operation; all 4 survived. Of 12 patients who experienced cardiogenic

Pathological Data Normal coronary arteries Inoperable coronary artery disease Extensive Single vessel Low ejection fraction (< 0.40) Raised SGOT Low EVR (< 0.70) Total

No. of Patients 0

4 2

2 2 1

11

Deaths Hosp

Late

. . .

. . .

0 1

1

0 0 0

0 0 2

0 0 0

IABP = intraaortic balloon pump assistmce; SGOT = serum glutamic oxaloacetic transaminase; EVR = endocardia1 viability ratio.

538 The Annals of Thoracic Surgery Vol 22 No 6 December 1976

Table 5. Significant Angiographic, Hemodynamic, and Biochemical Data in 31 Patients with Acute Coronary lnsufficiency Who Had lntraaortic Balloon Pump Assistance Preoperatively

Pathological Data

No. of

Patients

Clinical indications only 6 Low ejection fraction 12 (< 0.40) Left main coronary 7 artery stenosis or equivalent Low EVR (< 0.70) 5 Raised SGOT 1 Total 31

Deaths Hosp

Late

0 1

0 0

1

0

Although successfully resuscitated, she could not be weaned from cardiopulmonary bypass. The other patient died four weeks postoperatively, having remained in a low-output state after operation. The perioperative infarction rate was 19%. Myocardial infarction was proved by postmortem examination (3 nonsurvivors) or by the development of Q waves (4 survivors). There have been no late deaths in a follow-up period of 2 to 24 months (mean, 12 months).

Left Ventricular Dysfunction Of the 28 patients who were supported by IABP 0 0 because of left ventricular dysfunction, 18 had 0 0 coronary artery disease with both angina and 2 0 congestive heart failure and 10 had valvular disEVR = endocardial viability ratio; SGOT = serum glutarnic ease (Table 6). Other indicators of cardiac funcoxaloacetic transaminase. tion were also reduced; there was a decline in mean cardiac index in both groups and an inSerum glutamic oxaloacetic transaminase crease in left ventricular end-diastolic pressure (SGOT) levels greater than two times normal both prior to and after angiography. One of the were measured in 2 patients, and a low EVR was 18 patients with angina and congestive failure died six weeks postoperatively in a persistent present in another patient; all 3 survived. Another 31 patients with acute coronary in- low-output state despite successful weaning sufficiency had IABP prior to operation (see from IABP. Another died eight weeks postTable 5). Retrospective analysis revealed no ac- operatively of pulmonary embolism. Thus the ceptable angiographic or hemodynamic reason overall mortality in this group of high-risk pafor IABP to have been used in 6 patients. The tients has been 11%over a follow-up period of 3 remainder had one or more angiographic or to 21 months (mean, 10 months). Valve replacement in 10 patients was ashernodynamic reasons for counterpulsation: 12 had impairment of left ventricular function, 7 sociated with a 50% hospital mortality. These 10 had left main coronary artery stenosis, 5 had a patients represent 2.5% of our total valve relow EVR preoperatively, and 1 had an elevated placement experience during this period. There SGOT. Two patients died. One patient experi- was 1late death, giving a total mortality of 60% enced a cardiac arrest after general anesthesia over a follow-up period of 3 to 24 months (mean, was induced prior to insertion of the balloon. 11 months). Table 6 . Hernodynamic Data and Results of Operation in Patients with Left Ventricular Dysfunction" No. of

Mean LVEDP

Rest+Postangiog.

Cardiac Index

Deaths

Primary Disease

Patients

LVEF

(mm Hg)

(Limidm')

Coronary artery disease

18

19436

2.4

6

6

Valve disease

10

0.29 (0.08-0.40) 0.30 (0.10-0.40)

21442

2.2

50

10

"Defined as an ejection fraction less than 0.40.

LVEF = left ventricular ejection fraction; LVEDP = left ventricular end-diastolic pressure.

Hosp

(Oh)

Late

539 Gunstensen et al: Preoperative Intraaortic Balloon Pump Assist

Table 7 . Indication f o r Prophylactic lntraaortic Balloon Pump Assistance in Left Main Coronary Artery Disease Deaths Indications for IABP LMCA disease only indication LV dysfunction LMCA disease Low EVR + LMCA disease LV dysfunction + low EVR + LMCA disease Total

+

No. of Patients

Hosp

Late

12

0

0

6

1

0

1

0

0

1

0

0

20

1

0

IABP = intraaortic balloon pump assistance; LMCA = left main coronary artery; LV = left ventricle; EVR = endocardial viability ratio.

Left Main Coronary Artery Stenosis Twenty patients with stenosis of the left main coronary artery underwent operation (Table 7). The extent of coronary disease as revealed by angiography was the only indication for IABP in 12 patients. In 6, left ventricular dysfunction was an additional factor; a low EVR was present in 1, and another had both a low EVR and left ventricular dysfunction. The 1patient who died became hypertensive and developed ventricular tachycardia and then hypotension following induction of general anesthesia before IABP was started. Although he was returned to the recovery room, he remained in a low-output state until he died. There have been no late deaths in this group of patients, who have been followed for 2 to 22 months (mean, 7 months).

Endocardial Viability Ratio A retrospective analysis was made of the significance of the EVR (Table 8). The records of 24 patients who died after aortocoronary bypass in the years before IABP became available to us were retrospectively studied. The EVR in 11 (48%) of them was less than 0.70. These 24 patients were compared with 26 patients who went into cardiogenic shock after aortocoronary bypass and who were managed by IABP; 6 of the 26 who had IABP died, a 24% mortality. Half of this latter group had an EVR less than 0.70. Analysis of 50 randomly selected patients who survived aortocoronary bypass prior to the availability of IABP showed that only 6 (12%) had a low EVR. Extrapolating this experience, we have calculated that a low EVR implies a three- to fourfold increase in the risk of developing a postoperative low-output state after aortocoronary bypass grafting. Prophylactic IABP appears to abolish this increased risk, in that 18 of our patients who received balloon support for other reasons were found to have a low EVR, and they all had a smooth postoperative recovery and survived.

Complications from Use of IABP One death may have been due to the use of IABP. This patient could not be weaned from bypass after aortic valve replacement, and he died despite the use of IABP. He was subsequently shown to have aortic insufficiency-a contraindication to IABP. There has been 1loss of limb in a patient with cardiogenic shock after acute papillary muscle rupture. Four surviving patients have mild claudication, and 1required

Table 8 . Significance of Preoperative Endocardial Viability Ratio in Coronary Artery Disease a /'

with EVR

Survival ( %)

Patient Groups

Mean EVR

24 pts dead after ACB (no IABP)

0.78 0.82

50

0 24

0.98 0.55

12 100

100 100

< 0.70

~

26 pts with low cardiac output after ACB (postop IABP) 50 pts survivors of ACB (no IABP) 18 pts with low EVR (preop IABP)

48

EVR = endocardial viability ratio; ACB = aortocoronary bypass; IABP = intraaortic balloon pump assistance.

540 The Annals of Thoracic Surgery Vol 22 No 6 December 1976

arterioplasty to the common femoral artery. There have been no aortic dissections.

Comment Acute myocardial ischemia may be reversed by reduction of left ventricular ejection impedance, or afterload, which effects a decrease in myocardial oxygen requirements [31]. This may be achieved by mechanical circulatory assistance or by pharmacological means with administration of vasodilator agents [71 or P-adrenergic blocking agents [23]. Increases in myocardial oxygen supply may be effected by use of coronary vasodilators [21], but in critical coronary artery stenosis, flow may be dependent upon aortic diastolic blood pressure 111. The only means presently available to increase aortic diastolic blood pressure significantly while concurrently reducing myocardial oxygen requirement is diastolic counterpulsation. Permanent relief of ischemia, however, can only be obtained by myocardial revascularization. The timing of operation in relation to the acute ischemic syndromes of infarction and acute coronary insufficiency remains controversial [17, 22, 341. Clinical experience with early myocardial revascularization after infarction has been successful [20], but it is limited in practice to those patients who have sustained infarction while in the hospital. The hospital mortality after acute myocardial infarction is 15 to 20% and is principally related to left ventricular power failure [32]. Economics and logistics dictate that effort be directed toward these patients, who face an 80 to 90% mortality with conventional medical therapy. Intraaortic balloon assistance alone has resulted in improved short-term survival, but the longterm results are uncertain [5,9,331. Prompt cardiopulmonary bypass, cardiac catheterization, and appropriate operative procedures have significantly improved the results [171, but widespread use of this technique is not generally possible. Intraaortic balloon pumping allows stabilization of the patient’s hemodynamic status and permits safe cardiac catheterization [19]. The best results of treatment for shock have been observed when the potential for myocardial revascularization existed but the proper timing of corrective operation was as yet undetermined.

Logistics have forced an aggressive attitude upon us, and indeed, our only survivors have been in the group of patients who underwent immediate operation. They were also the patients who responded best to IABP, and it is likely that they would have survived operation undertaken at a later date as advocated by Mundth and co-workers [251. The early mortality after acute ventricular septa1 defect or papillary muscle rupture following myocardial infarction has been reported to be as high as 90% within two weeks [15]. Though survival after early repair has been recorded [6, 15, 301, most surgeons have favored delayed operation. The reduction in afterload made possible by IABP has significantly reduced the left-to-right shunting and degree of mitral regurgitation in these patients and has permitted useful stabilization of the circulation prior to operation [12]. Recurrent refractory ventricular arrhythmia early after myocardial infarction is universally associated with massive myocardial infarction in our experience [36]. Intraaortic balloon assistance was useful in supporting these patients but did not reduce the frequency of arrhythmias, a finding at variance with that reported by Mundth and colleagues [24]. Infarctectomy reduced the frequency of arrhythmia, which still persisted for as long as 10 days postoperatively in our experience. Balloon pumping was continued until the arrhythmia ceased. The presence of the balloon in the aorta was particularly useful when prolonged periods of external cardiac massage were required. The balloon was kept inflated during massage to direct cardiac output to the brain; in 1 patient, continuous external cardiac massage was maintained for 1 hour with eventual survival and normal cerebral function. Our survival rate of 6 out of 9 patients (67%) compares favorably with that reported by Mundth [24]. The ischemia of acute coronary insufficiency, which is severe but at the cellular level still nonlethal, may be present temporarily in as many as 10% of all patients with angina [38]. However, recurrent daily attacks of acute coronary insufficiency have been reported to be associated with a high mortality [lo]. Some authors have regarded the syndrome as a surgical emergency requiring immediate revascularization [22, 351, but considerable controversy still exists as to the

541 Gunstensen et al: Preoperative Intraaortic Balloon Pump Assist

best form of management [341. Our group had previously advocated operation after a period of medical therapy but abandoned this concept because of the high operative mortality [371. The most significant pathological finding in our series was subendocardial necrosis, and this was correlated with a low EVR preoperatively. We believe that recurrent acute coronary insufficiency eventually leads to subendocardial necrosis and fibrosis [37] that will adversely affect the results after delayed operation. Use of IABP increases the EVR and permits an aggressive surgical attitude toward these patients [ill. It has been advocated that IABP should be employed routinely to support the patient before angiography [ l l ] , but we have not found this necessary. However, our experience has led us to insist on initiating IABP under local anesthesia prior to induction of general anesthesia. Isom and associates [16] have shown that maximum creatine phosphokinase release occurs after induction of anesthesia but prior to cardiopulmonary bypass, which may explain why these patients tolerate anesthesia poorly. In some instances IABP was initiated several hours prior to operation to permit accommodation in a busy operating room schedule. This allows the privilege of time without the penalty of ongoing ischemic damage. Subendocardial ischemia, as evidenced by a low preoperative EVR, is not limited to patients with acute coronary insufficiency [14]; indeed, a low EVR was present in 50% of the patients who developed a low-output state after aortocoronary bypass. With prophylactic IABP, patients with a low preoperative EVR survived. We have previously suggested that a low preoperative EVR should be an indication for prophylactic IABP in patients about to undergo aortocoronary bypass [141. Patients with left main coronary artery stenosis face a high medical and surgical mortality [8, 391. The critical nature of the stenosis renders them particularly vulnerable to alterations in the myocardial oxygen supply/demand balance. Their brittle status is illustrated by 1 patient with critical stenosis of the sole patent artery in whom a rise in left ventricular afterload (systolic hypertension) after induction of anesthesia could not be met by a corresponding increase in blood supply. Ventricular tachycardia

resulted, followed by hypotension and then cardiac arrest. Reducing left ventricular afterload and increasing aortic diastolic pressure provides a considerable measure of safety for these patients. Since we began inserting the balloon catheter under local anesthesia prior to induction of general anesthesia, we have not had further problems in patients with left main coronary artery stenosis. Both short- and long-term survival after coronary artery revascularization or valve replacement is reduced in the presence of left ventricular dysfunction [181. Improved techniques of standard cardiopulmonary bypass have reduced myocardial injury during operation [4], but patients with left ventricular dysfunction and minimal cardiac reserve cannot afford to lose any more myocardial tissue. Improved organ function during cardiopulmonary bypass is provided by pulsatile flow systems [281, and IABP provides a simple method of introducing pulsatility to the standard cardiopulmonary bypass circuit [27]. Our results for patients with coronary artery disease have been encouraging, but the results achieved with valve replacement are less satisfactory. It is perhaps pertinent that the majority of patients requiring valve operations did not have coronary artery disease; their left ventricular dysfunction was more likely due to rheumatic myocardial fibrosis and terminal dilatation than to ischemia per se. Contrary to our initial expectations, the costeffectiveness of our IABP program has been very positive. In spite of the heavy use of balloon counterpulsation, the average length of stay for patients in our intensive care unit was reduced from 5.1 days in 1973 to 3.7 days in 1974. The total number of tracheostomies performed in 1973 was 13, compared with 3 in 1974. Renal dialysis was required on 10 occasions in 1973 but only twice in 1974. We believe these figures reflect the smooth and complication-free postoperative period afforded to critically ill patients by IABP.

References 1. Archie JP: Mechanical determinants of myocardial blood flow and its distribution. Ann Thorac Surg 20:39, 1975 2. Bregman D, Parodi EN, Edie RN, et al: Intraoperative unidirectional intra-aortic balloon pumping

542 The Annals of Thoracic Surgery

Vol 22 No 6 December 1976

in the management of left ventricular power failure. J Thorac Cardiovasc Surg 70:1010, 1975 3. Buckberg GD, Fixler DE, Archie JP, et al: Experimental subendocardial ischemia in dogs with normal coronary arteries. Circ Res 30:67, 1972 4. Buckberg GD, Olinger GN, Mulder DG, et al: Depressed postoperative cardiac performance: prevention by adequate myocardial protection during cardiopulmonary bypass. J Thorac Cardiovasc Surg 70:974, 1975 5. Buckley MJ, Leinbach RC, Kastor JA, et al: Hemodynamic evaluation of intra-aortic balloon pumping in man. Circulation 41:Suppl2:130, 1970 6. Buckley MJ, Mundth ED, Daggett WM, et al: Surgical management of ventricular septal defects and mitral regurgitation complicating acute myocardial infarction. Ann Thorac Surg 16:598, 1973 7. Chatterjee K, Swan HJC: Vasodilator therapy in acute myocardial infarction. Mod Concepts Cardiovasc Dis 43:119, 1974 8. Cohen MV, Cohn PF, Herman M V , et al: Diagnosis and prognosis of main left coronary artery obstruction. Circulation 45,46:Suppl 1:57, 1972 9. Dunkman WB, Leinbach RC, Buckley MJ, et al: Clinical and hemodynamic results of intra-aortic balloon pumping and surgery for cardiogenic shock. Circulation 46:465, 1972 10. Gazes PC, Mobley EM, Faris HM, et al: Preinfarctional (unstable) angina-a prospective study-ten year follow-up. Circulation 48:331, 1973 11. Gold HK, Leinbach RC, Sanders CA, et al: Intraaortic balloon pumping for control of recurrent myocardial ischemia. Circulation 47:1197, 1973 12. Gold HK, Leinbach RC, Sanders CA, et al: Intraaortic balloon pumping for ventricular septal defect or mitral regurgitation complicating acute myocardial infarction. Circulation 47:1191, 1973 13. Goldman BS, Gunstensen J, Gilbert BW, et al: Increasing operability and survival with intraaortic balloon pump assist. Can J Surg 19:69,1976 14. Gunstensen J, Scully HE, Kelly T, et al: The prognostic significance of the endocardia1 viability ratio (DPTIITTI) in aorto-coronary bypass surgery. Can J Surg 19:93, 1976 15. Iben AB, Miller RR, Amsterdam EA, et al: SUCcessful immediate repair of acquired ventricular septal defect and survival in patients with acute myocardial infarction shock using a new double patch technique. Chest 66:665, 1974 16. Isom OW, Spencer FC, Feigenbaum H, et al: Prebypass myocardial damage in patients undergoing coronary revascularization: an unrecognized vulnerable period. Circulation 52:Suppl 2:119, 1975 17. Keon WJ, Bedard P, Shankar KR, et al: Experience with emergency aorto-coronary bypass grafts in the presence of acute myocardial infarction. Circulation 47,48:Suppl3:151, 1973

18. Lefemine AA, Moon HS, Flessas A, et al: Myocardial resection and coronary artery bypass for left ventricular failure following myocardial infarction. Ann Thorac Surg 17:1, 1974 19. Leinbach RC, Dinsmore RE, Mundth ED, et al: Selective coronary and left ventricular cineangiography during intra-aortic balloon pumping for cardiogenic shock. Circulation 45:845, 1972 20. Loop FD, Cheanvechai C, Sheldon WC, et al: Early myocardial revascularization during acute myocardial infarction. Chest 65:478, 1974 21. Maroko PR, Braunwald, E. Modification of myocardial infarction size after coronary occlusion. Ann Intern Med 79:720, 1973 22. Matloff JM, Sustaita H, Chattejee K, et al: The rationale for surgery in pre-infarction angina. J Thorac Cardiovasc Surg 69:73, 1975 23. Mueller HS, Ayres SM, Religa A, et al: Propranolol in the treatment of acute myocardial infarction: effect on myocardial oxygenation and hemodynamics. Circulation 49:1078, 1974 24. Mundth ED, Buckley MJ, DeSanctis RW, et al: Surgical treatment of ventricular irritability. J Thorac Cardiovasc Surg 66:943, 1973 25. Mundth ED, Buckley MJ, Leinbach RC, et al: Myocardial revascularization for the treatment of cardiogenic shock complicating acute myocardial infarction. Surgery 70:78, 1971 26. O’Rourke MF, Chang VP, Windsor HM, et al: Acute severe cardiac failure complicating myocardial infarction. Br Heart J 37:169, 1974 27. Pappas G: A simple method of producing pulsatile flow during clinical cardiopulmonary bypass. Ann Thorac Surg 17:405, 1974 28. Pappas G, Winter SD, Kopriva CJ, et al: Improvement of myocardial and other vital organ functions and metabolism with a simple method of pulsatile flow (IABP) during clinical cardiopulmonary bypass. Surgery 7 7 3 , 1975 29. Philips PA, Marty AT, Miyamoto AM: A clinical method for detecting subendocardial ischemia after cardiopulmonary bypass. J Thorac Cardiovasc Surg 69:30, 1975 30. Pickering BN, Cleland WP: Ventricular septal defects due to acute myocardial infarction. J Cardiovasc Surg (Torino) 15:88, 1974 31. Sarnoff SJ, Braunwald E, Welch GH, et al: Hemodynamic determinants of oxygen consumption of heart with special reference to tension-time index. Am J Physiol 192:148, 1958 32. Scheidt S, Ascheim R, Killip T: Shock after acute myocardial infarction: a clinical and hemodynamic profile. Am J Cardiol26:556, 1970 33. Scheidt SS, Wilner G, Mueller H, et al: Intra-aortic balloon counterpulsation in cardiogenic shock: report of a co-operative clinical trial. N Engl J Med 288:979, 1973 34. Seldon R, Neil1 WA, Ritzman LW, et al: Medical versus surgical therapy for acute coronary insuffi-

543 Gunstensen et al: Preoperative Intraaortic Balloon Pump Assist

35. 36.

37.

38. 39.

ciency: a randomized study. N Engl J Med 293:1329, 1975 Spencer FC, Glassman E: Preinfarction angina: current therapeutic considerations. Am J Cardiol 32:382, 1973 Waxman MB, Wald RW, Goldman BS, et al: Intractable ventricular tachyarrhythmia post myocardial infarction. Circulation 51,52:Suppl2:109, 1975 Williams WG, Aldridge HE, Silver MD, et al: Preinfarction angina-what is it?, Coronary Artery Medicine and Surgery. Edited by JC Norman. New York: Appleton-Century-Crofts, 1975 Wood P: Acute and subacute coronary insufficiency. Br Med J 1:1779, 1961 Zeft HJ, Manley JC, Tector AJ, et al: Left main coronary artery stenosis: results of coronary bypass surgery. Circulation 49:68, 1974

Discussion

(Providence, RI): It has been a privilege to hear Dr. Gunstensen's thoughtful analysis. I will limit my remarks to the truly elective situation in which one inserts an intraaortic balloon preoperatively in patients undergoing coronary revascularization. Since we began our program of elective preoperative IABP for revascularization, we have inserted balloon catheters in 57 patients. Intraaortic balloons were inserted preoperatively in 26 patients with left main coronary artery stenosis (or the equivalent in severe ostial obstruction of both the left anterior descending and circumflex arteries) with no operative deaths. Twenty-six patients with unstable angina had IABP preoperatively with 2 deaths. The IABP was used preoperatively in an additional 5 patients with severe left ventricular dysfunction, with no deaths. In 9 of the patients the balloon was inserted before cardiac catheterization at the request of the cardiologists to stabilize hemodynamics. One of the deaths in the unstable angina group is worthy of comment in that i t occurred early in our experience and emphasizes two principles in the care of these patients. Propranolol was stopped preoperatively in this patient with intermediate coronary syndrome, and the balloon was inserted under local anesthesia before general anesthesia was induced in the operating room on the day of the operation. At that time the patient developed an atrial arrhythmia, tachycardia, hypertension, acute myocardial infarction, and cardiac arrest before the balloon could be inserted. On the basis of this case and similar considerations, we have now changed our thinking on propranolol administration and the time at which IABP is started. Elective preoperative balloon pumping is now started the day before rather than the day of operation to stabilize the patient's condition for some hours

before operation is undertaken. Propranolol now is continued until operation in all our patients to prevent arrhythmia, and if we have any concern about compromising ventricular function with propranolol, IABP obviates that risk. In our experience IABP relieves angina promptly and improves the preinfarction situation for catheterization and operation. The reduced coronary blood flow of left main coronary artery stenosis is protected against catastrophic compromise of hypotension by IABP. We therefore agree with Dr. Gunstensen's thesis, and we have become more and more dependent on IABP. We recommend preoperative balloon pumping to stabilize hemodynamics and improve the myocardial oxygen supply/ demand ratio. It seems reasonable, and our clinical experience has proved this to be true. I think we should continue to study this problem and evaluate further the indications for elective preoperative balloon counterpulsation. It appears to me, in line with what Dr. Gunstensen has pointed out, that IABP is a valuable adjunct.

DR. K A R L E. KARLSON

DR. DAVID BREGMAN (New York, NY): I compliment Dr. Gunstensen and his group on their eloquent presentation. It was particularly pleasing to see that a low EVR was employed as a criterion for IABP, a fact which correlates well with our own published data. During the past four years we have employed IABP during open-heart operations in 45 patients a t the Columbia-Presbyterian Medical Center. The patients were selected from more than 850 adult surgical patients. In this group of 45 patients 38 were weaned from bypass and from IABP, and 29, or 64%, were long-term survivors. The majority of these patients, of course, had IABP intraoperatively . In addition, we have recently successfully used balloon pumping in a 12-year-old child who was in a low-cardiac-output state after a complex congenital open-heart repair. A new approach in myocardial support during cardiopulmonary bypass has been suggested by Drs. Berger and Pappas. They created a pulsatile arterial pressure contour during cardiopulmonary bypass with an intraaortic balloon. We have developed a new and simple pulsatile assist device to convert roller pump flow to pulsatile flow. In addition, our pulsatile assist device can be used as an arterial counterpulsator before and after cardiopulmonary bypass. Our assist device, consisting.of a flexible valveless balloon through which the arterial blood flows, is inserted in the arterial line close to the aortic root. The balloon is contained within a rigid plastic housing connected to a standard intraaortic balloon pump. This device has been employed in 22 patients undergoing open-heart operation for coronary artery or aortic valve disease or both. Ten of these patients were in NYHA Functional Class IV, and 8 had ejection fractions of less than 0.4. The device functioned as a hemodynamically effective counterpulsator before

544 The Annals of Thoracic Surgery Vol 22 No 6 December 1976

and after cardiopulmonary bypass. During bypass pulse pressures of 40 to 50 mm Hg were readily obtained. Free plasma hemoglobin values after bypass were in the usual range for these patients. All survived, and there was no morbidity noted from use of the pulsatile assist device. No patient required IABP for weaning from cardiopulmonary bypass. The pulsatile assist device is a simple and reliable device, both for intraoperative arterio-arterial counterpulsation and for the creation of pulsatile cardiopulmonary bypass. In selected high-risk patients the pulsatile assist device appears to be a reasonable alternative to the elective use of IABP. More importantly, use of the pulsatile assist device may decrease the need for postoperative IABP. DR. ROQUE PIFARRE (Maywood, IL): I congratulate Dr. Gunstensen on his presentation and his results. I also support the principle that he has applied of preoperative use of balloon pumping. We have used IABP for patients with left main coronary artery disease. After operating on 89 such patients, we thought that the mortality rate of 11% was too high even for this lethal kind of disease, and since July, 1975, we have used preoperative IABP in23 patients without a death. Obviously the number of patients is very small, but I think the results are encouraging. We will continue the study in the hope that IABP will really contribute to a greater salvage of these patients. I also agree that IABP offers the advantage of safer induction of anesthesia. Pulsatile flow during the pump run is beneficial, as is IABP for postoperative support when indicated. DR. G . F R A N K 0.TYERS (Hershey, PA): I too congratulate Dr. Gunstensen and the Toronto group on their excellent report and would like to describe a potential additional indication for preoperative balloon support. In the absence of a fully developed picture of cardiogenic shock, we have used IABP and propranolol therapy to stabilize patients (including the Chief of Surgery) who have experienced recurrent or continued anginal pain following acute myocardial infarction. The value of this approach is supported by the frequent demonstration of almost normal ventricular function at subsequent precoronary bypass catheterization after what could have been a major myocardial infarction. We have now used balloon pumping in more than 125 patients. Of the 52 who received IABP preoperatively there were no 30-day operative deaths in the 8 who had it electively for poor ventricular function (ejection fraction < 0.30 in two projection planes) or left main coronary artery disease, whereas the 30-day mortality was 27% (12 patients) among the 44 who received emergency balloon support for preinfarction, decubitus, or postinfarction angina. In 40 pa-

tients with acquired heart disease who had an ejection fraction of less than 0.30, the 30-day mortality has been reduced from 25% to below 5% in the two years that we have been using IABP preoperatively in selected patients. While there have been reports of good survival in patients with low ejection fraction in the absence of balloon assistance, the fraction has most commonly been calculated from only the right anterior oblique projection, so that preoperative ventricular function was significantly poorer in our patients. Similarly, in more than 100 patients with left main coronary artery stenosis, we have noted a fall in the 30-day operative mortality from above 20% to less than 5% with selective preoperative IABP. The only recent death was that of a high-risk patient who, in addition to left main coronary artery stenosis, had aortoiliac disease that prevented prophylactic IABP. (Boston, MA): We at Boston University have been using IABP for more than six years and have employed it in approximately 100 patients. I am beginning to wonder if the pendulum is not swinging way, way over and if IABP is not being used excessively. The intraaortic balloon is not a totally innocuous device. It produces some morbidity and an occasional death. I have no disagreement with its application in patients in cardiogenic shock. The indications are clear-cut, and the temporary benefits are undisputed. The use of counterpulsation in unstable angina makes good sense. Actually, we alternated some patients in this category for a while but were unable to show any difference between those who had IABP and those who did not. Morbidity and mortality were essentially the same. It should be mentioned that if IABP is used in a patient with unstable angina, it should be deployed prior to cardiac catheterization for protection during a stressful period. It does not make much sense to take the patient through catheterization without support and then put the balloon in the next day. We have had no deaths in our last 25 patients who had bypass grafting for left main coronary artery obstruction without balloon support. This result obviously cannot be improved by counterpulsation. So I am not quite sure that we have a valid indication for IABP in this subset, and if anesthesia is carefully induced, there should be no major problems. We came up with one additional indication for the use of IABP in patients with cardiac disease who need general surgical procedures on an emergency basis. The group consists of patients in NYHA Functional Class 111 or IV and those who have had recent infarction and who require emergency noncardiac operations. We used IABP in 5 such patients, and the results were quite good. Finally, I would like to ask one question: Do the authors have any hemodynamic data on the use of DR. ROBERT L. BERGER

545 Gunstensen et al: Preoperative Intraaortic Balloon Pump Assist

IABP in patients not in shock? We have some information, and it is quite interesting. (Wayland, MA): We have all seen some very impressive data today. I would like to present our own experience of 92 applications in 90 patients. We first used IABP prophylactically in a patient with poor ventricular function about five years ago. Our indications for using IABP agree in most respects with those of the other discussants. These include inability to wean from the pump because of postoperative shock and shock in the intensive care area postoperatively. In patients weaned from the pump, survival was 47%. When counterpulsation was applied postoperatively, survival was 20%. Seventeen patients have had prophylactic balloon support. All 17 were in NYHA Functional Class IV. They had poor ventricular function and were undergoing resection of the ventricular wall for adynamic segments or ventricular aneurysm. In this group we have had a 94% survival. We have also used IABP for preoperative shock in 4 patients, with 2 survivors. Postinfarction angina was the indication in 1of these patients. Eleven patients (73%) survived among the 15 treated for cardiogenic shock without additional operation. We had a very high rate of unsuccessful passage of the balloon catheter in the medical group (20°/0), particularly if they were over age 70. There are problems with the balloon. A certain number cannot be passed through the femoral artery (9%), and a certain number have serious vascular obstruction problems leading to muscle damage (1 patient) or gangrene of the leg (2 patients); this has been more true of patients with cardiogenic shock than of those with postoperative shock. We have also

had to remove the balloon on at least 2 occasions postoperatively because of severe vasoconstriction that restricted circulation to the leg.

D R . ARMAND A . LEFEMINE

We agree with Dr. Karlson that propranolol withdrawal can be a problem, and after having been caught a couple of times earlier in this series, we now continue propranolol administration right up to the time of operation. Similarly, I think it is useful to insert the balloon perhaps 1or 2 days before operation in patients with severe preinfarction angina. This approach allows the patient to be accommodated in a busy operating room schedule without suffering further myocardial damage in the interim. I think Dr. Bregman’s comments about pulsatile flow are very well taken. But I wonder whether it is as useful in high-risk patients as inserting the intraaortic balloon prior to induction of general anesthesia, which I think is the high-risk time for these patients. It is difficult to know if we are using the balloon excessively. We employ it in a high percentage of patients, but because of our use of the balloon, we have increased the number of severely ill patients who are referred to our hospital. We utilize it for preinfarction angina only if medical therapy has failed, and we do not use it for patients who have passed through one or two attacks of acute coronary insufficiency. I agree with Dr. Lefemine that the procedure is not entirely without complications. In 90 elective cases, however, we have had only 4 patients who experienced minor degrees of claudication postoperatively, and just 1 of them required arterioplasty. A single serious complication, gangrene of the leg, occurred in a patient with severe cardiogenic shock. We have not had any deaths directly attributable to the use of IABP. D R . GUNSTENSEN: