639
JACC Vol. 19, No.3 March I, 1992:639-46
REPORTS ON THERAPY
Percutaneous Transluminal Coronary Angioplasty in Patients With Cardiogenic Shock MICHAEL D. HIBBARD, MD, DAVID R. HOLMES, JR., MD, FACC, KENT R. BAILEY, PHD, GUY S. REEDER, MD, FACC, JOHN F. BRESNAHAN, MD, FACC, BERNARD J. GERSH, MB,CHB, DPHIL, FACC Rochester, Minnesota
In cardiogenic shock complicating acute myocardial infarction, percutaneous transluminal coronary angioplasty has been reported to significantly improve the modest survival benefits afforded by emergency surgical revascularization and thrombolytic therapy. The records of all patients who underwent angioplasty for acute myocardial infarction complicated by cardiogenic shock were retrospectively reviewed to determine whether coronary angioplasty improves survival. Of the 45 patients, 28 (group 1, 62%) had successful dilation of the infarct-related artery and 17 (group 2, 38%) had unsuccessful angioplasty. The groups were similar in extent of coronary artery disease, infarct location, incidence of multivessel disease and hemodynamic variables. The overall hospital survival rate was 56% (71 % in group 1 and 29% in group 2). Group 1 patients had
more left main coronary artery disease, and group 2 patients were older and had a higher incidence of prior myocardial infarction. Multivariate analysis showed that the survival advantage in patients with successful angioplasty was statistically significant (p 0.014) when these factors were taken into account. At a mean follow-up interval of 2.3 years (range 1 month to 5.6 years), there were five deaths (four cardiac and one noncardiac), for a 2.3-year survival rate of 80% in patients surviving to hospital discharge. During the follow-up period, 36% of hospital survivors had repeat hospitalization for cardiac evaluation, 8% had myocardial infarction, 8% had coronary artery bypass surgery and 24% had angina. (J Am Coli CardioI1992j19:639-46)
Despite improvements in coronary care, cardiogenic shock remains a lethal complication of acute myocardial infarction 0-6). Therapeutic strategies that did not include an attempt to reestablish perfusion of the infarct-related artery have yielded disappointing results, with hospital mortality rates of 80% to 90% (7-12). Even with thrombolytic therapy, survival has been poor. In the Gruppo Italiano per 10 Studio della Streptochinasi nell infarto Miocardico (GISSI) study (3) of 146 patients in Killip class IV, the mortality rate was 70%, similar to the 67% rate reported with intracoronary streptokinase in the registry of the Society for Cardiac Angiography (14). Recently, it was suggested 05-18) that angioplasty may reduce the mortality associated with cardiogenic shock complicating acute myocardial infarction. If these reports can be confirmed in larger series, angioplasty may prove to be the only realistic initial therapy for patients with cardiogenic shock. The Mayo Clinic experience was reviewed
retrospectively to examine the effect of coronary angioplasty on the in-hospital and long-term survival of patients with myocardial infarction complicated by cardiogenic shock,
From the Division of Cardiovascular Diseases and Internal Medicine and the Section of Biostatistics. Mayo Clinic and Mayo Foundation. Rochester. Minnesota. This work was presented in part at the 63rd Annual Scientific Session of the American Heart Association. Dallas, Texas, November 1990. Manuscript received January 23, 1991; revised manuscript received July 8, 1991, accepted September 11, 1991. Address for reprints: David R. Holmes, Jr.. MD, Mayo Clinic, 200 First Street S.W., Rochester. Minnesota 55905. © 1992 by the American College of Cardiology
=
Methods Study patients. The charts of all patients seen at the Mayo Clinic between October 1982 and November 1989 who underwent coronary angioplasty for cardiogenic shock or acute myocardial infarction within 48 h of admission were reviewed. The patients with myocardial infarction complicated by cardiogenic shock form the study group. The diagnosis of myocardial infarction was supported by I) chest pain in a clinical syndrome consistent with myocardial ischemia, 2) typical increase in creatine kinase MB isoenzyme activity, and 3) an electrocardiogram (ECG) with 2::2 mm of ST elevation in two consecutive leads in a typical ischemic syndrome. Cardiogenic shock was defined as one or more of these three findings: decreased urine output, cool extremities or changes in mental status in patients with 1) systolic blood pressure ::::95 mm Hg without inotropic or intraaortic balloon pump support, 2) systolic blood pressure ::::110 mm Hg with intraaortic balloon pump or inotropic support, or 3) evidence of adequate volume expansion demonstrated by measured filling pressures or pulmonary edema. Six patients were 0735-1097/92/$5.00
640
lACC Vol. 19. No.3 March I. 1992:639-46
HIBBARD ET AL. ANGIOPLASTY IN CARDIOGENIC SHOCK
undergoing cardiopulmonary resuscitation at the time of catheterization, and adequate hemodynamic measurements were not possible. Patients who had other possible explanations for shock, such as bleeding and sepsis, were excluded. Patients were included only if they had had angioplasty within 24 h of the onset of symptoms. All patients meeting these criteria were included regardless of age. Of the 291 patients reviewed, 57 met study criteria for cardiogenic shock. Seven patients were excluded because their medical records revealed evidence of significant bleeding, infection or malignancy that may have contributed to their hemodynamic derangement. Three patients were excluded because they had unstable angina and no enzymatic evidence of infarction. Two other patients were excluded; one had only a transient decrease in blood pressure after reperfusion with recombinant tissue-type plasminogen activator (rt-PA) and the other did not have dilation of the infarct-related artery. The remaining 45 patients form the study group. Patients were treated with medications, mechanical circulatory support and coronary bypass surgery at the discretion of the attending cardiologist. The infarct-related artery was identified by angiographic documentation of thrombus within the artery or by an occluded artery in the distribution of the patient's EeG changes of infarction. In three patients vessels in addition to the infarct-related artery were dilated. Coronaryangioplasty. After a diagnostic coronary angiogram, angioplasty was carried out as previously described (19,20). All patients received 5,000 to 15,000 U of heparin by intravenous bolus during angioplasty and then a heparin infusion to maintain the activated partial thromboplastin time at approximately 1.5 times the control value. Nitrates were given either parenterally or sublingually for ischemia. Aspirin was used in most (62%) of the patients. The patients were subsequently divided into two groups based on the success of angioplasty. Group I consisted of 28 patients with successful angioplasty (>20% improvement in angiographic stenosis as judged by visual estimate). Group 2 consisted of 17 patients in whom angioplasty was unsuccessful. Follow-up. Follow-up data were obtained by telephone contact with each patient or next of kin. When follow-up events such as myocardial infarction or coronary artery bypass surgery occurred at an outside institution, the relevant medical records were reviewed after consent was obtained from the hospital and patient or next of kin. Data analysis. Because this was an uncontrolled study of coronary angioplasty in cardiogenic shock, an attempt was made to compare baseline characteristics of the two groups and adjust for these in the analysis. Baseline variables were compared between group 1 and group 2 by use of the Pearson chi-square test for dichotomous variables and the two-sample t test for continuous or scale variables. Hospital mortality was compared between groups 1 and 2 by use of Pearson's chi-square test. This unadjusted comparison was followed with logistic models for mortality outcome, with reperfusion status, and with other baseline variables used as
Table 1. Clinical Characteristics of 45 Patients With Cardiogenic
Shock Complicating Acute Myocardial Infarction* Successful PTCA (group 1)
Clinical data Procedure Gender Male Female Cardiac risk factors Smoking (currently or formerly) Hypertension Diabetes Medical history Prior Mit Prior CABG or PTCA Angina 6 mo before angioplasty
Unsuccessful PTCA (group 2)
No.
%
No.
%
28
62
17
38
17 11
61 39
12 5
71 29
20 11 5
71 39 18
14 8 2
82 47 12
5 4 8
18 14 29
8 1 5
47 6 29
'Mean age (±SD) was 60 ± ILl years (range 41 to 79) in group I and 66.9 ± 12.3 years (range 41 to 86) in group 2. tStatistically significant difference (p = 0.036); other variables were not significantly different. CABG = coronary artery bypass graft; MI = myocardial infarction; PTCA = percutaneous transluminal coronary angioplasty.
independent predictor variables. Baseline variables were considered for entry into the models whether or not they had a statistically significant relation to outcome, because many real associations would not be statistically detectable in a study of this size.
Results Patient characteristics (Table 1). Gender, history of smoking, hypertension. diabetes and angina were not significantly different between group 1 and group 2. Group 2 patients were slightly older than group 1patients (66.9 years vs. 60 years; p = 0.055). More patients in group 2 than in group 1had a history of prior myocardial infarction (47% vs. 18%; p = 0.036). Overall, 71 % of patients required intraaortic balloon pump support. 49% had ventricular tachycardia or ventricular fibrillation requiring emergency cardioversion, 29% required a temporary pacemaker, 22% had documented right ventricular infarction and 24% of hospital survivors required cardiopulmonary resuscitation. Hemodynamic data (Table 2). There were no statistically significant differences between the groups in heart rate, systolic blood pressure or filling pressure. Hemodynamic data could not be obtained for six patients who were undergoing cardiopulmonary resuscitation at the time of angioplasty. Two patients in group 1did not have an invasive assessment of left ventricular filling pressure but were included because of hypotension in association with clinical and radiologic evidence of pulmonary edema. Extent of coronary disease (Table 3). There was no statistically significant difference in the location of the infarct-
JACC Vol. 19. No.3 March I. 1992:639-46
Table 2. Hemodynamic Data in the 45 Study Patients* Successful PTCA (group I) No. of Patients Hemodynamic variable Heart rate (beats/min) Systolic BP (mm Hg) Filling pressures (mm Hg)t LV end-diastolic Wedge Pulmonary artery diastolic
Mean ± SD
Unsuccessful PICA (group 2) No. of Patients
28 28
88 ± 33 78 ± 14
17 17
95 ± 26 78 ± 12
16 5 2
27 ± 10 20 ± 9 20 ± 2
7 6
25 ± 8 23 ± 8 15
I
related artery between group 1and group 2. More patients in group 1 than in group 2 had single-vessel disease (39% vs. 18%), but the difference was not statistically significant (p = 0.128). More patients in group 2 than in group 1 had two-vessel disease (35% vs. 4%; p = 0.004). The two groups had a similar proportion of patients with three-vessel disease (43% in group 1 and 41% in group 2). Four patients in group 1 and one patient in group 2 had left main coronary artery disease. Ten patients had right ventricular infarction documented by echocardiography, by radionuclide angiography, at operation or at autopsy. Coronary angioplasty (Table 4). Overall, 62% of attempted dilations were successful (group 1) and 38% were Table 3. Extent of Coronary Artery Disease in 45 Patients Successful PICA (group I)
Table 4. Morphologic Features and Success of Angioplasty in 43 of the 45 Study Patients* Successful PICA (n = 27)
Mean ± SD
*No differences were statistically significant. t Adequate hemodynamic measurements were not possible in three patients in group I and three in group 2 because they were undergoing cardiopulmonary resuscitation at the time of catheterization. Two patients in group I did not have invasive assessment of filling pressure but were included because of clinical and radiographic evidence of pulmonary edema. BP = blood pressure; LV = left ventricular: other abbreviation as in Table I.
Vessel dilated RCA LAD LCx Vein graft to OM Left main artery Coronary artery disease* Single-vessel disease Two-vessel diseaset Three-vessel disease Left main disease Infarction Documented RV infarct
641
HIBBARD ET AL. ANGIOPLASTY IN CARDIOGENIC SHOCK
Unsuccessful PICA (group 2)
No.
%
No.
%
10 14 3
36 50 II
4 II 2
2
7
24 65 12 6 6
II I 12 4
39 4 43 14
3 6 7 I
18 35 41 6
7
25
18
*~ 70% visual estimate stenosis in the left anterior descending (LAD), left circumflex (LCx) and right coronary arteries (RCA): >50% estimate stenosis in the left main coronary artery. tSignificantly different. p = 0.004; other variables were not significantly different. OM = obtuse marginal artery; RV = right ventricular; other abbreviation as in Table I.
Total occlusion Total or subtotal Stenosis ~50% Distal to lesion Not visible Branch involved Stenosis characteristIcs Calcium ~mild Tortuosity ~mild Bend ~45° Length ~lOmm
Cannot quantitate Thrombus ~5 mm Cannot quantitate Dissection present Collateral vessels Distal embolization
Unsuccessful PICA (n = 16)
Overall 43)
(n =
No.
%
No.
%
No.
%
22 27
81 100
12 15
75 94
34 42
79 98
15 I 7
56 4 26
7 3 5
44 19 31
22 4 12
51 9 28
24 25 2
89 93 7
14 14 I
88 88 6
38 39 3
88 91 7
5 10
19 37
I 7
6 44
6 17
14 40
7 13 5 7 2
26 48 19 26 7
I II 2 I I
6 69 13 6 6
8 24 7 8 3
19 56 16 19 7
*Two angiograms could not be located for review. No differences were statistically sigmficant. Abbreviation as in Table I.
unsuccessful (group 2). There was a trend for the elderly to have a lower angioplasty success rate; only 6 (43%) of 14 patients>70 years old had a successful procedure compared with 22 (71%) of 31 patients 70 old had been performed during or after 1988. The coronary angiograms from 43 of the 45 patients were available for review by two experienced angiographers who looked for morphologic features historically related to angioplasty success. There were no statistically significant differences in lesion morphology between patients with successful or unsuccessful procedures. Overall, 98% of 43 culprit lesions were totally or subtotally occluded, 51 % had stenosis 2::50% in the distal vessel and 28% had significant branch vessel involvement. Culprit lesion calcification and tortuosity were mild or absent in 88% and 91% of lesions, respectively. Fewer than 7% of the lesions were located on an arterial curve 2::45 0 , and there were no ostial lesions. It was not possible to adequately quantitate lesion length and approximate volume of thrombus because the distal vessel could not be visualized in a substantial portion of the angiograms reviewed. Among the 17 patients with unsuccessful dilation, in 11 (64%) the failure was due to inability to cross the lesion, in 3 (18%) to the persistent reaccumulation of thrombus and in 3
642
IACC Vol. 19, No.3 March I, 1992:639-46
HIBBARD ET AL. ANGIOPLASTY IN CARDIOGENIC SHOCK
Table 5. Hospital Management of the 45 Study Patients' Successful PTCA (group I)
Mechanical intervention IABP VT or VF cardioversion Temporary pacemaker Repeat PTCA Hospital events Emergency CABG Elective CABG Reinfarction
Unsuccessful PTCA (group 2)
No.
%
No.
%
19 13 6 It
68 46 21 4
13
9 7
76 53 41
5 4 4
18 14 14
2 2 2:j:
12 12 12
'No differences were statistically significant. tThe infarct-related artery was reocduded; repeat dilation was unsuccessful. :j:One reinfarction was fatal. IABP = intraaortic balloon pump; VF = ventricular fibrillation; VT = ventricular tachycardia; other abbreviations as in Table I.
(18%) to cardiac arrest and death during the procedure after sustained attempts at revascularization. The mean time from the onset of symptoms to angioplasty was 5.7 ± 4.6 h (range 0.5 to 21.5) if dilation was successful and 6.6 ± 6.5 h (range 1.5 to 22.5) if it was unsuccessful. Patients who had successful dilation had a higher peak creatine kinase MB level (211 compared with 137 IV/liter), but the difference was not statistically significant. Hospital management (Table 5). There was no significant difference in the use of dopamine, dobutamine, digoxin, beta-adrenergic blocking agents, calcium channel blocking agents, nitrates, nitroprusside, lidocaine, heparin or thrombolytic agents between the two groups. Thrombolytic agents were used in 32% and 24% of patients in group 1 and group 2, respectively. They were administered as primary treatment in 8 of 13 patients (6 in group I and 2 in group 2). Cardiogenic shock later developed in these patients and angioplasty was attempted as secondary therapy. In one patient (group 2), intracoronary streptokinase was administered after sustained attempts to cross the culprit lesion had failed. The remaining four patients (three in group I, one in group 2) received a thrombolytic agent for residual thrombus after primary angioplasty. Significantly more patients in group 2 than in group I received epinephrine (71% vs. 39%; p = 0.042). Group 2 patients were also significantly less likely than group 1 patients to receive aspirin (35% vs. 79%; p = 0.004). The two groups were similar in frequency of intraaortic balloon pump support and coronary bypass surgery. More patients in group 2 than in group 1 required temporary pacemaker placement (41% vs. 21%), but the difference was not significant (p = 0.156). Before hospital discharge, seven patients underwent emergency surgery and six underwent elective surgical revascularization. The five patients with successful angioplasty who required emergency bypass surgery all had hemodynamic stabilization before bypass surgery; four had severe three-vessel disease and one was operated on after
dilation of the left main coronary artery. One patient with previously successful angioplasty had repeat catheterization for recurrent chest pain. At angiography, the infarct-related artery had reoccluded, and attempts at repeat dilation were unsuccessful. There was no significant difference between group I and group 2 in the incidence of in-hospital reinfarction. One reinfarction in group 2 was fatal. In-hospital mortality. Overall 25 (56%) of the 45 patients survived to hospital discharge. The survival rate was 71% in patients with successful angioplasty compared with 29% in patients with unsuccessful dilation (p = 0.006). To obtain an adjusted estimate of the effect of successful angioplasty, a forward stepwise multiple logistic regression analysis was performed, with hospital mortality as the outcome and baseline variables-including age, gender, left main coronary artery disease, time from onset of symptoms to dilation, filling pressures, number of diseased vessels, vessel dilated, prior myocardial infarction and requirement for cardiopulmonary resuscitation-as the potential predictors of survival. With this model, independent predictors of in-hospital mortality were the success of angioplasty, the presence or absence of left main coronary artery disease, age and the need for cardiopulmonary resuscitation before angioplasty. Successful angioplasty resulted in a significant negative contribution to mortality risk (p = 0.021), with an adjusted odds ratio of 11.6 in favor of successful angioplasty (95% confidence intervallA to 98.0). In an analysis of combined early and late survival after hospital discharge, there was a nonsignificant trend toward improved survival with successful angioplasty. There was no significant difference in survival between men and women. Age was an important determinant of survival. Of the 14 patients aged ;:::70 years, only 2 (14%) survived to hospital discharge compared with 23 (74%) of 31 patients
JACC Vol. 19, No.3 March I, 1992:639-46
REPORTS ON THERAPY
Percutaneous Transluminal Coronary Angioplasty in Patients With Cardiogenic Shock MICHAEL D. HIBBARD, MD, DAVID R. HOLMES, JR., MD, FACC, KENT R. BAILEY, PHD, GUY S. REEDER, MD, FACC, JOHN F. BRESNAHAN, MD, FACC, BERNARD J. GERSH, MB,CHB, DPHIL, FACC Rochester, Minnesota
In cardiogenic shock complicating acute myocardial infarction, percutaneous transluminal coronary angioplasty has been reported to significantly improve the modest survival benefits afforded by emergency surgical revascularization and thrombolytic therapy. The records of all patients who underwent angioplasty for acute myocardial infarction complicated by cardiogenic shock were retrospectively reviewed to determine whether coronary angioplasty improves survival. Of the 45 patients, 28 (group 1, 62%) had successful dilation of the infarct-related artery and 17 (group 2, 38%) had unsuccessful angioplasty. The groups were similar in extent of coronary artery disease, infarct location, incidence of multivessel disease and hemodynamic variables. The overall hospital survival rate was 56% (71 % in group 1 and 29% in group 2). Group 1 patients had
more left main coronary artery disease, and group 2 patients were older and had a higher incidence of prior myocardial infarction. Multivariate analysis showed that the survival advantage in patients with successful angioplasty was statistically significant (p 0.014) when these factors were taken into account. At a mean follow-up interval of 2.3 years (range 1 month to 5.6 years), there were five deaths (four cardiac and one noncardiac), for a 2.3-year survival rate of 80% in patients surviving to hospital discharge. During the follow-up period, 36% of hospital survivors had repeat hospitalization for cardiac evaluation, 8% had myocardial infarction, 8% had coronary artery bypass surgery and 24% had angina. (J Am Coli CardioI1992j19:639-46)
Despite improvements in coronary care, cardiogenic shock remains a lethal complication of acute myocardial infarction 0-6). Therapeutic strategies that did not include an attempt to reestablish perfusion of the infarct-related artery have yielded disappointing results, with hospital mortality rates of 80% to 90% (7-12). Even with thrombolytic therapy, survival has been poor. In the Gruppo Italiano per 10 Studio della Streptochinasi nell infarto Miocardico (GISSI) study (3) of 146 patients in Killip class IV, the mortality rate was 70%, similar to the 67% rate reported with intracoronary streptokinase in the registry of the Society for Cardiac Angiography (14). Recently, it was suggested 05-18) that angioplasty may reduce the mortality associated with cardiogenic shock complicating acute myocardial infarction. If these reports can be confirmed in larger series, angioplasty may prove to be the only realistic initial therapy for patients with cardiogenic shock. The Mayo Clinic experience was reviewed
retrospectively to examine the effect of coronary angioplasty on the in-hospital and long-term survival of patients with myocardial infarction complicated by cardiogenic shock,
From the Division of Cardiovascular Diseases and Internal Medicine and the Section of Biostatistics. Mayo Clinic and Mayo Foundation. Rochester. Minnesota. This work was presented in part at the 63rd Annual Scientific Session of the American Heart Association. Dallas, Texas, November 1990. Manuscript received January 23, 1991; revised manuscript received July 8, 1991, accepted September 11, 1991. Address for reprints: David R. Holmes, Jr.. MD, Mayo Clinic, 200 First Street S.W., Rochester. Minnesota 55905. © 1992 by the American College of Cardiology
=
Methods Study patients. The charts of all patients seen at the Mayo Clinic between October 1982 and November 1989 who underwent coronary angioplasty for cardiogenic shock or acute myocardial infarction within 48 h of admission were reviewed. The patients with myocardial infarction complicated by cardiogenic shock form the study group. The diagnosis of myocardial infarction was supported by I) chest pain in a clinical syndrome consistent with myocardial ischemia, 2) typical increase in creatine kinase MB isoenzyme activity, and 3) an electrocardiogram (ECG) with 2::2 mm of ST elevation in two consecutive leads in a typical ischemic syndrome. Cardiogenic shock was defined as one or more of these three findings: decreased urine output, cool extremities or changes in mental status in patients with 1) systolic blood pressure ::::95 mm Hg without inotropic or intraaortic balloon pump support, 2) systolic blood pressure ::::110 mm Hg with intraaortic balloon pump or inotropic support, or 3) evidence of adequate volume expansion demonstrated by measured filling pressures or pulmonary edema. Six patients were 0735-1097/92/$5.00
640
lACC Vol. 19. No.3 March I. 1992:639-46
HIBBARD ET AL. ANGIOPLASTY IN CARDIOGENIC SHOCK
undergoing cardiopulmonary resuscitation at the time of catheterization, and adequate hemodynamic measurements were not possible. Patients who had other possible explanations for shock, such as bleeding and sepsis, were excluded. Patients were included only if they had had angioplasty within 24 h of the onset of symptoms. All patients meeting these criteria were included regardless of age. Of the 291 patients reviewed, 57 met study criteria for cardiogenic shock. Seven patients were excluded because their medical records revealed evidence of significant bleeding, infection or malignancy that may have contributed to their hemodynamic derangement. Three patients were excluded because they had unstable angina and no enzymatic evidence of infarction. Two other patients were excluded; one had only a transient decrease in blood pressure after reperfusion with recombinant tissue-type plasminogen activator (rt-PA) and the other did not have dilation of the infarct-related artery. The remaining 45 patients form the study group. Patients were treated with medications, mechanical circulatory support and coronary bypass surgery at the discretion of the attending cardiologist. The infarct-related artery was identified by angiographic documentation of thrombus within the artery or by an occluded artery in the distribution of the patient's EeG changes of infarction. In three patients vessels in addition to the infarct-related artery were dilated. Coronaryangioplasty. After a diagnostic coronary angiogram, angioplasty was carried out as previously described (19,20). All patients received 5,000 to 15,000 U of heparin by intravenous bolus during angioplasty and then a heparin infusion to maintain the activated partial thromboplastin time at approximately 1.5 times the control value. Nitrates were given either parenterally or sublingually for ischemia. Aspirin was used in most (62%) of the patients. The patients were subsequently divided into two groups based on the success of angioplasty. Group I consisted of 28 patients with successful angioplasty (>20% improvement in angiographic stenosis as judged by visual estimate). Group 2 consisted of 17 patients in whom angioplasty was unsuccessful. Follow-up. Follow-up data were obtained by telephone contact with each patient or next of kin. When follow-up events such as myocardial infarction or coronary artery bypass surgery occurred at an outside institution, the relevant medical records were reviewed after consent was obtained from the hospital and patient or next of kin. Data analysis. Because this was an uncontrolled study of coronary angioplasty in cardiogenic shock, an attempt was made to compare baseline characteristics of the two groups and adjust for these in the analysis. Baseline variables were compared between group 1 and group 2 by use of the Pearson chi-square test for dichotomous variables and the two-sample t test for continuous or scale variables. Hospital mortality was compared between groups 1 and 2 by use of Pearson's chi-square test. This unadjusted comparison was followed with logistic models for mortality outcome, with reperfusion status, and with other baseline variables used as
Table 1. Clinical Characteristics of 45 Patients With Cardiogenic
Shock Complicating Acute Myocardial Infarction* Successful PTCA (group 1)
Clinical data Procedure Gender Male Female Cardiac risk factors Smoking (currently or formerly) Hypertension Diabetes Medical history Prior Mit Prior CABG or PTCA Angina 6 mo before angioplasty
Unsuccessful PTCA (group 2)
No.
%
No.
%
28
62
17
38
17 11
61 39
12 5
71 29
20 11 5
71 39 18
14 8 2
82 47 12
5 4 8
18 14 29
8 1 5
47 6 29
'Mean age (±SD) was 60 ± ILl years (range 41 to 79) in group I and 66.9 ± 12.3 years (range 41 to 86) in group 2. tStatistically significant difference (p = 0.036); other variables were not significantly different. CABG = coronary artery bypass graft; MI = myocardial infarction; PTCA = percutaneous transluminal coronary angioplasty.
independent predictor variables. Baseline variables were considered for entry into the models whether or not they had a statistically significant relation to outcome, because many real associations would not be statistically detectable in a study of this size.
Results Patient characteristics (Table 1). Gender, history of smoking, hypertension. diabetes and angina were not significantly different between group 1 and group 2. Group 2 patients were slightly older than group 1patients (66.9 years vs. 60 years; p = 0.055). More patients in group 2 than in group 1had a history of prior myocardial infarction (47% vs. 18%; p = 0.036). Overall, 71 % of patients required intraaortic balloon pump support. 49% had ventricular tachycardia or ventricular fibrillation requiring emergency cardioversion, 29% required a temporary pacemaker, 22% had documented right ventricular infarction and 24% of hospital survivors required cardiopulmonary resuscitation. Hemodynamic data (Table 2). There were no statistically significant differences between the groups in heart rate, systolic blood pressure or filling pressure. Hemodynamic data could not be obtained for six patients who were undergoing cardiopulmonary resuscitation at the time of angioplasty. Two patients in group 1did not have an invasive assessment of left ventricular filling pressure but were included because of hypotension in association with clinical and radiologic evidence of pulmonary edema. Extent of coronary disease (Table 3). There was no statistically significant difference in the location of the infarct-
JACC Vol. 19. No.3 March I. 1992:639-46
Table 2. Hemodynamic Data in the 45 Study Patients* Successful PTCA (group I) No. of Patients Hemodynamic variable Heart rate (beats/min) Systolic BP (mm Hg) Filling pressures (mm Hg)t LV end-diastolic Wedge Pulmonary artery diastolic
Mean ± SD
Unsuccessful PICA (group 2) No. of Patients
28 28
88 ± 33 78 ± 14
17 17
95 ± 26 78 ± 12
16 5 2
27 ± 10 20 ± 9 20 ± 2
7 6
25 ± 8 23 ± 8 15
I
related artery between group 1and group 2. More patients in group 1 than in group 2 had single-vessel disease (39% vs. 18%), but the difference was not statistically significant (p = 0.128). More patients in group 2 than in group 1 had two-vessel disease (35% vs. 4%; p = 0.004). The two groups had a similar proportion of patients with three-vessel disease (43% in group 1 and 41% in group 2). Four patients in group 1 and one patient in group 2 had left main coronary artery disease. Ten patients had right ventricular infarction documented by echocardiography, by radionuclide angiography, at operation or at autopsy. Coronary angioplasty (Table 4). Overall, 62% of attempted dilations were successful (group 1) and 38% were Table 3. Extent of Coronary Artery Disease in 45 Patients Successful PICA (group I)
Table 4. Morphologic Features and Success of Angioplasty in 43 of the 45 Study Patients* Successful PICA (n = 27)
Mean ± SD
*No differences were statistically significant. t Adequate hemodynamic measurements were not possible in three patients in group I and three in group 2 because they were undergoing cardiopulmonary resuscitation at the time of catheterization. Two patients in group I did not have invasive assessment of filling pressure but were included because of clinical and radiographic evidence of pulmonary edema. BP = blood pressure; LV = left ventricular: other abbreviation as in Table I.
Vessel dilated RCA LAD LCx Vein graft to OM Left main artery Coronary artery disease* Single-vessel disease Two-vessel diseaset Three-vessel disease Left main disease Infarction Documented RV infarct
641
HIBBARD ET AL. ANGIOPLASTY IN CARDIOGENIC SHOCK
Unsuccessful PICA (group 2)
No.
%
No.
%
10 14 3
36 50 II
4 II 2
2
7
24 65 12 6 6
II I 12 4
39 4 43 14
3 6 7 I
18 35 41 6
7
25
18
*~ 70% visual estimate stenosis in the left anterior descending (LAD), left circumflex (LCx) and right coronary arteries (RCA): >50% estimate stenosis in the left main coronary artery. tSignificantly different. p = 0.004; other variables were not significantly different. OM = obtuse marginal artery; RV = right ventricular; other abbreviation as in Table I.
Total occlusion Total or subtotal Stenosis ~50% Distal to lesion Not visible Branch involved Stenosis characteristIcs Calcium ~mild Tortuosity ~mild Bend ~45° Length ~lOmm
Cannot quantitate Thrombus ~5 mm Cannot quantitate Dissection present Collateral vessels Distal embolization
Unsuccessful PICA (n = 16)
Overall 43)
(n =
No.
%
No.
%
No.
%
22 27
81 100
12 15
75 94
34 42
79 98
15 I 7
56 4 26
7 3 5
44 19 31
22 4 12
51 9 28
24 25 2
89 93 7
14 14 I
88 88 6
38 39 3
88 91 7
5 10
19 37
I 7
6 44
6 17
14 40
7 13 5 7 2
26 48 19 26 7
I II 2 I I
6 69 13 6 6
8 24 7 8 3
19 56 16 19 7
*Two angiograms could not be located for review. No differences were statistically sigmficant. Abbreviation as in Table I.
unsuccessful (group 2). There was a trend for the elderly to have a lower angioplasty success rate; only 6 (43%) of 14 patients>70 years old had a successful procedure compared with 22 (71%) of 31 patients 70 old had been performed during or after 1988. The coronary angiograms from 43 of the 45 patients were available for review by two experienced angiographers who looked for morphologic features historically related to angioplasty success. There were no statistically significant differences in lesion morphology between patients with successful or unsuccessful procedures. Overall, 98% of 43 culprit lesions were totally or subtotally occluded, 51 % had stenosis 2::50% in the distal vessel and 28% had significant branch vessel involvement. Culprit lesion calcification and tortuosity were mild or absent in 88% and 91% of lesions, respectively. Fewer than 7% of the lesions were located on an arterial curve 2::45 0 , and there were no ostial lesions. It was not possible to adequately quantitate lesion length and approximate volume of thrombus because the distal vessel could not be visualized in a substantial portion of the angiograms reviewed. Among the 17 patients with unsuccessful dilation, in 11 (64%) the failure was due to inability to cross the lesion, in 3 (18%) to the persistent reaccumulation of thrombus and in 3
642
IACC Vol. 19, No.3 March I, 1992:639-46
HIBBARD ET AL. ANGIOPLASTY IN CARDIOGENIC SHOCK
Table 5. Hospital Management of the 45 Study Patients' Successful PTCA (group I)
Mechanical intervention IABP VT or VF cardioversion Temporary pacemaker Repeat PTCA Hospital events Emergency CABG Elective CABG Reinfarction
Unsuccessful PTCA (group 2)
No.
%
No.
%
19 13 6 It
68 46 21 4
13
9 7
76 53 41
5 4 4
18 14 14
2 2 2:j:
12 12 12
'No differences were statistically significant. tThe infarct-related artery was reocduded; repeat dilation was unsuccessful. :j:One reinfarction was fatal. IABP = intraaortic balloon pump; VF = ventricular fibrillation; VT = ventricular tachycardia; other abbreviations as in Table I.
(18%) to cardiac arrest and death during the procedure after sustained attempts at revascularization. The mean time from the onset of symptoms to angioplasty was 5.7 ± 4.6 h (range 0.5 to 21.5) if dilation was successful and 6.6 ± 6.5 h (range 1.5 to 22.5) if it was unsuccessful. Patients who had successful dilation had a higher peak creatine kinase MB level (211 compared with 137 IV/liter), but the difference was not statistically significant. Hospital management (Table 5). There was no significant difference in the use of dopamine, dobutamine, digoxin, beta-adrenergic blocking agents, calcium channel blocking agents, nitrates, nitroprusside, lidocaine, heparin or thrombolytic agents between the two groups. Thrombolytic agents were used in 32% and 24% of patients in group 1 and group 2, respectively. They were administered as primary treatment in 8 of 13 patients (6 in group I and 2 in group 2). Cardiogenic shock later developed in these patients and angioplasty was attempted as secondary therapy. In one patient (group 2), intracoronary streptokinase was administered after sustained attempts to cross the culprit lesion had failed. The remaining four patients (three in group I, one in group 2) received a thrombolytic agent for residual thrombus after primary angioplasty. Significantly more patients in group 2 than in group I received epinephrine (71% vs. 39%; p = 0.042). Group 2 patients were also significantly less likely than group 1 patients to receive aspirin (35% vs. 79%; p = 0.004). The two groups were similar in frequency of intraaortic balloon pump support and coronary bypass surgery. More patients in group 2 than in group 1 required temporary pacemaker placement (41% vs. 21%), but the difference was not significant (p = 0.156). Before hospital discharge, seven patients underwent emergency surgery and six underwent elective surgical revascularization. The five patients with successful angioplasty who required emergency bypass surgery all had hemodynamic stabilization before bypass surgery; four had severe three-vessel disease and one was operated on after
dilation of the left main coronary artery. One patient with previously successful angioplasty had repeat catheterization for recurrent chest pain. At angiography, the infarct-related artery had reoccluded, and attempts at repeat dilation were unsuccessful. There was no significant difference between group I and group 2 in the incidence of in-hospital reinfarction. One reinfarction in group 2 was fatal. In-hospital mortality. Overall 25 (56%) of the 45 patients survived to hospital discharge. The survival rate was 71% in patients with successful angioplasty compared with 29% in patients with unsuccessful dilation (p = 0.006). To obtain an adjusted estimate of the effect of successful angioplasty, a forward stepwise multiple logistic regression analysis was performed, with hospital mortality as the outcome and baseline variables-including age, gender, left main coronary artery disease, time from onset of symptoms to dilation, filling pressures, number of diseased vessels, vessel dilated, prior myocardial infarction and requirement for cardiopulmonary resuscitation-as the potential predictors of survival. With this model, independent predictors of in-hospital mortality were the success of angioplasty, the presence or absence of left main coronary artery disease, age and the need for cardiopulmonary resuscitation before angioplasty. Successful angioplasty resulted in a significant negative contribution to mortality risk (p = 0.021), with an adjusted odds ratio of 11.6 in favor of successful angioplasty (95% confidence intervallA to 98.0). In an analysis of combined early and late survival after hospital discharge, there was a nonsignificant trend toward improved survival with successful angioplasty. There was no significant difference in survival between men and women. Age was an important determinant of survival. Of the 14 patients aged ;:::70 years, only 2 (14%) survived to hospital discharge compared with 23 (74%) of 31 patients
