A randomized controlled trial of allopurinol coronary bypass surgery
in
A plethora of experimental evidence indicates that allopurinol reduces the formation of cytotoxic free radicals during myocardial ischemia and reperfusion. The purpose of this study was to evaluate the effect of allopurinol on cardiac performance and early mortality after coronary bypass surgery. Allopurinol (n = 89) or placebo (n = 80) was administered to 189 patients before surgery. Randomization produced groups evenly matched for surgical risk factors. Hospital mortality rate in the placebo group was 14 of 80 (18%) in the allopurinol group 4 of 89 (4%), p = 0.014. Cardiac performance, scored by cardiac index and the need for ionotropic or mechanical support, was significantly better in the allopurinol group. More nonfatal complications occurred in the allopurinol group. When either a complication or death is termed an event, the proportion of events was equal in the two groups. No side effects were identified. We now administer allopurinol to all patients who are undergoing bypass surgery unless specifically contraindicated. (AM HEART J 1991;121:20.)
W. Dudley Johnson, MD, Kenneth L. Kayser, MS, Jerold B. Brenowitz, Saed F. Saedi, MD, Milwaukee, Wis.
The purpose of this study was to determine whether pretreatment with allopurinol would alter cardiac function after coronary bypass surgery. A plethora of theoretical and experimental evidence suggests that allopurinol protects various tissues from reperfusion injury following ischemia. This study was concerned with the effect of allopurinol pretreatment on myocardium that was rendered temporarily ischemic during coronary bypass surgery. A substantial body of relevant knowledge is recorded in the literature. Simpson et a.l.l present an excellent summary. Various experimental studies demonstrate improved function in ischemic animal hearts pretreated with allopurinol.2-8 Other studies have shown reduced infarct size in various experimental preparations.g-*2 Reduction of arrhythmias13 and reduction of creatinine kinase release following ischemia3, I4 have also been reported. Two groups15y l6 have reported negative results on infarct size. The mechanism of protection is not completely known; however, excellent reviews have been published.l, 17-lg In essence, ischemia causes the conversion of xanthine dehydrogenase to xanthine oxiFrom St. Mary’s Medical Center, Milwaukee, Wis. Received for publication June 4, 1990; accepted July 24, 1990. Reprint requests: W. Dudley Johnson, MD, 2315 North Lake Drive, 1007, Milwaukee, WI 53211. 4i1124853
20
Suite
MD, and
dase. Hypoxanthine accumulates in the cell. When molecular oxygen reappears, xanthine oxidase acts on substrates oxygen and hypoxanthine to form the superoxide anion radical, -02, hydrogen peroxide, H202, and the hydroxyl radical, -OH. These free radicals are cytotoxic. Other mechanisms for the formation of free radicals have been proposed,l> “9 lg but recent research20* 21 indicates that the reaction described above is the main source of free radicals. Efforts at thwarting the ill effects of free radicals center on preventing their formation or scavenging the radicals after they are formed. Allopurinol works by metabolically blocking their formation. Allopurinol is rapidly converted to its metabolite, oxypurinol, which then binds to xanthine oxidase and inhibits its action. The scavenger, superoxide dismutase, converts superoxide anion to Hz02; catalase then converts Hz02 to water. In this article we are concerned only with the action of orally administered allopurino1 on myocardium that is rendered temporarily ischemic during coronary bypass surgery. In spite of the huge amount of experimental evidence, we have found only one report on the use of allopurinol in human beings with myocardial ischemia.22 This group reported that patients undergoing cardiac surgery who were pretreated with allopurinol had a much higher incidence of spontaneous reversion to sinus rhythm and fewer metabolic changes than untreated patients.
Volume 121 Number 1, Part 1
Table
Randomized
I. Reason for exculsion*
Allopurinol
Number Characteristics
Refused to participate Preop allopurinol for gout Surgeon’s option Emergency Schedule change Medication error Contraindication-renal Apprehensive Valve replacement planned *Ninety-three
patients
30 16 13 13 11 6 2 1 1
of 262 who were operated
on.
METHODS Study design.
After approval by the human subjects committee, the study commenced on November 16, 1987. Every patient scheduled to undergo bypass surgery was offered the opportunity to be included in the study. The only exceptions were patients scheduled to have concomitant valve replacement. The protocol specified 200 subjects and periodic review. After informed consent was obtained, each study subject received two doses of “study drug” (either allopurinol or placebo) according to weight: under 66 kg = 200 mg X2,66 to 93 kg = 300 mg X2, and over 93 kg = 400 mg X2. The first dose was given the evening before surgery and the second, 4 hours before scheduled surgery time. If the medication schedule could not be attained, the subject was dropped from the study. Everyone associated with the study itself or with care of the patient was blinded to the identity of the drug. We did break the code on several occasions when a side effect of allopurinol was suspected. No side effects were identified. Effectiveness of the drug was evaluated by cardiac function in the 48 hours immediately after surgery and by the overall hospital mortality rate. Cardiac function was assessed by careful examination of hand-recorded and computerized (Hewlett-Packard Patient Data Management System, Hewlett-Packard Co., Palo Alto, Calif.) intensive ‘care records. One person classified all patients without knowledge of which drug was given and assigned an “adequacy score” as follows: Score 1: cardiac index over 2.0 without any support; Score 2: pharmacologic or mechanical (balloon pump) support required to maintain index over 2.0; Score 3: pharmacologic and mechanical support required to maintain index over 2.0 or prolonged, high doses of ionotropic drugs required to keep index in the range of 2.0 to 2.2; and Score 4: cardiac index under 2.0 in spite of maximal pharmacologic and mechanical support. If a patient had low cardiac output because of hypovolemia, a low index
was disregarded
until
adequate
preload
(pul-
monary artery wedge or central venous pressure over 18 mm Hg) was attained. Surgical technique. Our surgical technique has been centered on the goal of achieving complete revascularization. The mammary artery is used whenever practical. A vessel not amenable to conventional bypass is reconstructed with extensive endarterectomy with subsequent
in CABG
21
II. Preoperative characteristics
Table
Reason
trial of allopurinol
(n = 89)
Diabetes Insulin Oral medication None Pulmonary disease* Angina? None 1 2 3 4 Unknown Angina medicationt None 1 Med 2 Meds 3 Meds Previous CNS$ Female Vessel1 disease l-vessel 2-vessel 3-vessel Renal dysfunctioql Previous coronary surgery 0 1 2 or more Previous PTCA 0 1 2 or more Heart size Normal Enlarged Ejection fraction (% ) 20% 20-29s 30-39s 40-49 % 5o+Yc Average rt_SD Age b-r) 34-49 50-59 60-69 70-76 Average *Abnormal
pulmonary
Placebo (n = 80)
71 5
17 8 55 11
9 12 5 16 45 2
10 12 8 9 41 0
3 13 43 30 12 16
2 21 37 20 11 11
2 24 63 22
2 17 61 23
45 35 9
40 35 5
11 9 3
68 5 7
49 40
42 38
2 6 9 22 50 48.5 f 13
2 11 5 20 42 46.5 2 14
15 34 26 14 59.1 -r- 9.8
7 21 41
9 9
61.3
11 2 8.0
function, emphysema, asthma, chronic obstructive
pulmonary disease. tCanadian Cardiovascular Society classification. $Nitrates, @-blockers, or calcium antagonists. §Carotid disease, stroke, transient ischemic attack. I/Blood urea nitrogen over 20 or creatinine over 1.3.
long anastomosis. 22Perfusion was high flow, high pressure (target values 70 ml/kg and 60 mm Hg mean arterial pressure) with total body hypothermia of 30’ C to 34’ C. Intermittent ischemic arrest was used exclusively with 25
22
Johson
Table
et al.
III. Surgical
American
Table
details Allopurinol (n = 89)
Procedures
Total grafts 1 2 3 4 5 6 I 8 Average Mammary grafts 0 1 2 Average End grafts* 1 2 3 4+ Average Aneurysm resection Mitral valve replacement Total ischemic time Total pump time *Graft
Table
to vessel after
2 3 22 18
11
11
Table
50 35 4 0.48 k 0.59
53 23 4 0.39 t 0.58
26 20 12
23 14 3 2 0.87 k 1.11 7 1 131.7 -+ 50.2 243.8 l?r 83.7
5 3 130.2 + 46.6 235.1 ? 89.1
Score
1.40
1 2 3 4
score Allopurinol (n = 89)
Placebo (n = 80)
54 24 9 2 = 7.32; 3 degrees
of freedom.
4189 14 of 80
4.5(‘rs 17.5%
VI. Causes of death Group
4 1 4.3 li-
1
Allopurinol Placebo
Rate
p
in
A plethora of experimental evidence indicates that allopurinol reduces the formation of cytotoxic free radicals during myocardial ischemia and reperfusion. The purpose of this study was to evaluate the effect of allopurinol on cardiac performance and early mortality after coronary bypass surgery. Allopurinol (n = 89) or placebo (n = 80) was administered to 189 patients before surgery. Randomization produced groups evenly matched for surgical risk factors. Hospital mortality rate in the placebo group was 14 of 80 (18%) in the allopurinol group 4 of 89 (4%), p = 0.014. Cardiac performance, scored by cardiac index and the need for ionotropic or mechanical support, was significantly better in the allopurinol group. More nonfatal complications occurred in the allopurinol group. When either a complication or death is termed an event, the proportion of events was equal in the two groups. No side effects were identified. We now administer allopurinol to all patients who are undergoing bypass surgery unless specifically contraindicated. (AM HEART J 1991;121:20.)
W. Dudley Johnson, MD, Kenneth L. Kayser, MS, Jerold B. Brenowitz, Saed F. Saedi, MD, Milwaukee, Wis.
The purpose of this study was to determine whether pretreatment with allopurinol would alter cardiac function after coronary bypass surgery. A plethora of theoretical and experimental evidence suggests that allopurinol protects various tissues from reperfusion injury following ischemia. This study was concerned with the effect of allopurinol pretreatment on myocardium that was rendered temporarily ischemic during coronary bypass surgery. A substantial body of relevant knowledge is recorded in the literature. Simpson et a.l.l present an excellent summary. Various experimental studies demonstrate improved function in ischemic animal hearts pretreated with allopurinol.2-8 Other studies have shown reduced infarct size in various experimental preparations.g-*2 Reduction of arrhythmias13 and reduction of creatinine kinase release following ischemia3, I4 have also been reported. Two groups15y l6 have reported negative results on infarct size. The mechanism of protection is not completely known; however, excellent reviews have been published.l, 17-lg In essence, ischemia causes the conversion of xanthine dehydrogenase to xanthine oxiFrom St. Mary’s Medical Center, Milwaukee, Wis. Received for publication June 4, 1990; accepted July 24, 1990. Reprint requests: W. Dudley Johnson, MD, 2315 North Lake Drive, 1007, Milwaukee, WI 53211. 4i1124853
20
Suite
MD, and
dase. Hypoxanthine accumulates in the cell. When molecular oxygen reappears, xanthine oxidase acts on substrates oxygen and hypoxanthine to form the superoxide anion radical, -02, hydrogen peroxide, H202, and the hydroxyl radical, -OH. These free radicals are cytotoxic. Other mechanisms for the formation of free radicals have been proposed,l> “9 lg but recent research20* 21 indicates that the reaction described above is the main source of free radicals. Efforts at thwarting the ill effects of free radicals center on preventing their formation or scavenging the radicals after they are formed. Allopurinol works by metabolically blocking their formation. Allopurinol is rapidly converted to its metabolite, oxypurinol, which then binds to xanthine oxidase and inhibits its action. The scavenger, superoxide dismutase, converts superoxide anion to Hz02; catalase then converts Hz02 to water. In this article we are concerned only with the action of orally administered allopurino1 on myocardium that is rendered temporarily ischemic during coronary bypass surgery. In spite of the huge amount of experimental evidence, we have found only one report on the use of allopurinol in human beings with myocardial ischemia.22 This group reported that patients undergoing cardiac surgery who were pretreated with allopurinol had a much higher incidence of spontaneous reversion to sinus rhythm and fewer metabolic changes than untreated patients.
Volume 121 Number 1, Part 1
Table
Randomized
I. Reason for exculsion*
Allopurinol
Number Characteristics
Refused to participate Preop allopurinol for gout Surgeon’s option Emergency Schedule change Medication error Contraindication-renal Apprehensive Valve replacement planned *Ninety-three
patients
30 16 13 13 11 6 2 1 1
of 262 who were operated
on.
METHODS Study design.
After approval by the human subjects committee, the study commenced on November 16, 1987. Every patient scheduled to undergo bypass surgery was offered the opportunity to be included in the study. The only exceptions were patients scheduled to have concomitant valve replacement. The protocol specified 200 subjects and periodic review. After informed consent was obtained, each study subject received two doses of “study drug” (either allopurinol or placebo) according to weight: under 66 kg = 200 mg X2,66 to 93 kg = 300 mg X2, and over 93 kg = 400 mg X2. The first dose was given the evening before surgery and the second, 4 hours before scheduled surgery time. If the medication schedule could not be attained, the subject was dropped from the study. Everyone associated with the study itself or with care of the patient was blinded to the identity of the drug. We did break the code on several occasions when a side effect of allopurinol was suspected. No side effects were identified. Effectiveness of the drug was evaluated by cardiac function in the 48 hours immediately after surgery and by the overall hospital mortality rate. Cardiac function was assessed by careful examination of hand-recorded and computerized (Hewlett-Packard Patient Data Management System, Hewlett-Packard Co., Palo Alto, Calif.) intensive ‘care records. One person classified all patients without knowledge of which drug was given and assigned an “adequacy score” as follows: Score 1: cardiac index over 2.0 without any support; Score 2: pharmacologic or mechanical (balloon pump) support required to maintain index over 2.0; Score 3: pharmacologic and mechanical support required to maintain index over 2.0 or prolonged, high doses of ionotropic drugs required to keep index in the range of 2.0 to 2.2; and Score 4: cardiac index under 2.0 in spite of maximal pharmacologic and mechanical support. If a patient had low cardiac output because of hypovolemia, a low index
was disregarded
until
adequate
preload
(pul-
monary artery wedge or central venous pressure over 18 mm Hg) was attained. Surgical technique. Our surgical technique has been centered on the goal of achieving complete revascularization. The mammary artery is used whenever practical. A vessel not amenable to conventional bypass is reconstructed with extensive endarterectomy with subsequent
in CABG
21
II. Preoperative characteristics
Table
Reason
trial of allopurinol
(n = 89)
Diabetes Insulin Oral medication None Pulmonary disease* Angina? None 1 2 3 4 Unknown Angina medicationt None 1 Med 2 Meds 3 Meds Previous CNS$ Female Vessel1 disease l-vessel 2-vessel 3-vessel Renal dysfunctioql Previous coronary surgery 0 1 2 or more Previous PTCA 0 1 2 or more Heart size Normal Enlarged Ejection fraction (% ) 20% 20-29s 30-39s 40-49 % 5o+Yc Average rt_SD Age b-r) 34-49 50-59 60-69 70-76 Average *Abnormal
pulmonary
Placebo (n = 80)
71 5
17 8 55 11
9 12 5 16 45 2
10 12 8 9 41 0
3 13 43 30 12 16
2 21 37 20 11 11
2 24 63 22
2 17 61 23
45 35 9
40 35 5
11 9 3
68 5 7
49 40
42 38
2 6 9 22 50 48.5 f 13
2 11 5 20 42 46.5 2 14
15 34 26 14 59.1 -r- 9.8
7 21 41
9 9
61.3
11 2 8.0
function, emphysema, asthma, chronic obstructive
pulmonary disease. tCanadian Cardiovascular Society classification. $Nitrates, @-blockers, or calcium antagonists. §Carotid disease, stroke, transient ischemic attack. I/Blood urea nitrogen over 20 or creatinine over 1.3.
long anastomosis. 22Perfusion was high flow, high pressure (target values 70 ml/kg and 60 mm Hg mean arterial pressure) with total body hypothermia of 30’ C to 34’ C. Intermittent ischemic arrest was used exclusively with 25
22
Johson
Table
et al.
III. Surgical
American
Table
details Allopurinol (n = 89)
Procedures
Total grafts 1 2 3 4 5 6 I 8 Average Mammary grafts 0 1 2 Average End grafts* 1 2 3 4+ Average Aneurysm resection Mitral valve replacement Total ischemic time Total pump time *Graft
Table
to vessel after
2 3 22 18
11
11
Table
50 35 4 0.48 k 0.59
53 23 4 0.39 t 0.58
26 20 12
23 14 3 2 0.87 k 1.11 7 1 131.7 -+ 50.2 243.8 l?r 83.7
5 3 130.2 + 46.6 235.1 ? 89.1
Score
1.40
1 2 3 4
score Allopurinol (n = 89)
Placebo (n = 80)
54 24 9 2 = 7.32; 3 degrees
of freedom.
4189 14 of 80
4.5(‘rs 17.5%
VI. Causes of death Group
4 1 4.3 li-
1
Allopurinol Placebo
Rate
p
