- Email: [email protected]
Clinical implications of cigarette smoking in acute myocardial infarction: Acute angiographic findings and long-term prognosis
Clinical implications of cigarette smoking in acute myocardial infarction: Acute angiographic findings and long-term prognosis M a s a h a r u I s h i h a r a , MD, H i k a r u Sato, MD, H i r o n o b u Tateishi, MD, Takuji Kawagoe, MD, Yuji S h i m a t a n i , MD, S a t o s h i K u r i s u , MD, K a z u k o Saka|, MD, a n d K e n t a r o u Ueda, MD Hiroshima, Japan
This study was undertaken to assesswhether reperfusion in smokers could be achieved spontaneously or therapeutically and to assesswhether favorable outcome in smokers could be sustained for years after infarction. We studied 260 patients with anterior myocardial infarction who underwent coronary angiography and thrombolysis within 24 hours after the onset of chest pain. There were 158 smokers and 102 nonsmokers. Smoking was associated more with men, younger age, and less multivessel disease. On initial angiography, the distribution of Thrombolysis in Myocardial Infarction grade was similar between smokers and nonsmokers. After thrombolysis, Thrombolysis in Myocardial Infarction grade 3 was more frequent in smokers (32% vs 18%; p = 0.004). In-hospital mortality rates were lower (8% vs 18%; p = 0.022) and long-term cardiac survival was better in smokers (5-year survival: 82% vs 70%; p = 0.022). Our data demonstrated that the infarct artery of smokers respgnded more efficiently to thrombotysis and favorable outcome in smokers was sustained throughout 5 years. (Am HeartJ 1997;134:955-60.)
It has been: well established that cigarette smoking is a risk factor for acute myocardial infarction and s u d d e n cardiac d e a t h ) ,2 Paradoxically, most studies have reported that, in patients w h o have had acute myocardial infarction, a smoking habit before acute myocardial infarction is a predictor of favorable short-term outcome after acute myocardial infarction. 3-1~ Recent studies, reviewing coronary angiography early after thrombolytic therapy, demonstrated that a patent infarct artery is more frequent in smokers and suggested m ' o possibilities: commonness of spontaneous reperfusion and efficient response of occlusive coronary thrombus to thrombolytic therapy. 8q2 Because of lack of angiograp h y before thrombolytic therapy, however, it remained unclear whether reperfusion in smokers was achieved spontaneously or therapeutically. To assess this issue, w e reviewed coronary angiograms that were performed early after acute myocardial infarction but before the initiation of thrombolytic therapy. In addition, although previous studies have demonstrated favorable shortterm outcome after acute myocardial infarction in smokers, it has not b e e n addressed whether favorable outcome in smokers is sustained for years. Because Fromthe Deportmentof Cardiobgy, HiroshimaCity Hospital Submitted Feb. 25, 1997; acceptedJune 17, 199Z Reprint requests:Masaharu Ishihara,MD, Departmentof Cardiology, HiroshimaCity Hospital 7-33, Moto-machi, Naka-ku, Hiroshima, 730 Japan. Copyright 9 1997 by Mosby-Year Book, Inc. 0002-8703/97/$5.00 + 0 411184850
smoking is a risk factor for acute myocardial infarction and sudden cardiac death, smokers may b e more likely to suffer cardiac death years after hospital discharge. We also assessed the effect of smoking on long-term prognosis in relation to acute angiographic findings.
Methods Study
patients
This study consisted of 260 consecutive patients with anterior wall acute myocardial infarction who underwent coronary angiography and thrombolytic therapy within 24 hours after the onset of chest pain between January 1981 and December 1990. Anterior wall acute myocardial infarction was diagnosed by chest pain consistent with ongoing myocardial ischemia persisting longer than 30 minutes, at least 1 mm ST segment elevation in at least m'o adjacent precordial electrocardiographic leads, and elevation of serum creatine kinase levels beyond twice the normal upper limit. Thrombolytic therapy performed after coronary angiography identified the left anterior descending artery as the infarct artery.
Cardiac catheterization Emergent cardiac catheterization was performed through the right femoral artery after heparin administration. Contrast left ventriculography was performed in the 30degree right anterior oblique projection. Selective coronary angiography was performed in multiple projections. Thrombolytic therapy was performed with urokinase. Usually, cumulative doses of urokinase were infused up to 96 x l0 t IU until reperfusion was achieved. Adjunctive coronary angioplasty was performed in 121 (47%) patients.
956
Ishihara et al.
Figure I 100
80
. . . . . . .
........
,
,.' 9176176176176 ..'.~ .o . . . o- .. o." ..~ 9 9 ~
.o ..- .." ~ . . " , o9 .~ .~ .- .. ~ ~176. ~ . . " . ,
".."~176176 .. . ...~ .~ .o . ~ ..~ .~1769176.., .- . . . ~ .- ..
' ..- ~176176176176 . . . . . ,.." ..'...'.~ . . . . .
9ITtMtO.- ."..".-".."..-'. .- . . . . . . . -
Incidence
6O
(%)
9 ........ o~ ~ ~ j 9" . . . ' 9 1 7 6 1 7 6
9
, . . . . . ~ ..~ . , ' . , ,, ..- ..- ,.~ , . " .
9-'...'..~149176176 ' . . , ' . . . ' . . . ~ 1 4 9
40
20
-ili-i~i-~i.~.~ :(1.2~.~.~i( 9 .. ~ ..... .o .o , . . . . . . . . . . .~ ~176.." . 9 . . . . , .
'~176149 .., .- .~ .~ ..... ~ ~ . . . - . . , ~176. " .
,~711@1,1N\ "- - - - " 9 : : ~ :
[:!:!:i:i:[:i:i:~[[!~:i:i:i:i:i:[:i:~:i
i
Smokers Non-smokers Before thrombolysis
Smokers Non-smokers After thrombolysis
TIMI flow grade before and after thrombolytic therapy. Left, Before thrombolytic therapy, distribution of TIMI flow grade was similar between smokers and nonsmokers (p = 0.936). Right, After thrombolytic therapy, there was tendency toward more frequent reperfusion in smokers compared with nonsmokers (69% vs 59%; p = 0.099). TIMI flow grade 3 was significantly more frequent in smokers after thrombolytic therapy (32% vs 18%; p = 0.004).
The allocation of thrombolysis or angioplasty was not randomized and was b a s e d on the physician's decision.
Angiographic analysis All coronary angiograms were reviewed by two angiographers without knowledge of the clinical variables. The peffusion status of the left anterior descending artery was determined in accordance with the Thrombolysis in Myocardial Infarction (TIMI) study classification. 13 Initial TIMI flow grade was assessed before the initiation of thrombolytic therapy. Efficiency of thrombolytic therapy was assessed on the angiogram after the final dose of urokinase was administered: 30 to 60 minutes after the initiation of urokinase. Final TIMI flow grade was assessed on the final shot of the angiography. An occluded artery was defined as TIMI flow grade <1. Reperfusion was defined as TIMI flow grade >_2. Multivessel coronary disease was defined as >_75% stenosis in one or more vessels remote from the infarct artery. Left main coronary disease was considered to be at least two-vessel involvement. Collateral circulation was considered to b e present w h e n there was partial or complete filling of the epicardial segment of the left anterior descending coronary artery. The 30-degree right anterior oblique end-diastolic and end-systolic left ventricular endocardial contours from sinus beats were traced by a single angiographer. The left ventricular ejection fraction was calculated acc9rding to the area-length method, t4
or clinical visit. Data for a full 5 years were available in most (233) patients. The primary end point was cardiac death defined as sudden death, death after acute myocardial infarction, or death after hospitalization for congestive heart failure9
Data analysis Statistical analysis was performed with the Z2 and t tests9 Kaplan-Meier estimates were used to construct a long-term su~,ival curve. Differences in long-term survival were assessed with the log-rank test. Cox proportional hazards regression was used to identify independent predictors of long-term survival. Differences were considered significant if p < 0.05. All group data are expressed as mean + SD.
Results Baseline clinical characteristics This study group included 260 patients. Of 260 patients, 158 (61%) had a smoking habit at the time of acute myocardial infarction (smokers) and 102 (39~ did not (nonsmokers) (Table I). Average tobacco consumption was 23 +_ 15 cigarettes per day in smokers. Smokers were more likely to be male and younger compared with nonsmokers. There was no difference in hypertension, diabetes, previous infarction, antecedent angina, or Killip class on admission to the study.
Follow-up
Initial angiographic findings
Follow-up was conducted between January 1995 a n d April 1995. Follow-up was achieved for 252 (97%) patients by mail
Average time from the onset of acute myocardial infarction to emergency angiography was 5.3 -+ 4.4
American Heart Journd November 1997
Ishihara et al.
Table h Baseline clinical and initial ongiographic findings
Age (yr) Men {n/%) Hypertension (n/%) Diabetes (n/%) Previous infarction (n/%) Antecedent angina (n/%) Killlp class (n/%) 1 2 3 4 Time to angiography (hr) Initial TIMI flow grade 0 I 2 3 I Collateral circulation (n/%) No. diseased Vessels {n/%) 1 2 3 LVEF (%}
Smokers (n = 158)
Nonsmokers (n = 102)
57+ 11 146/92
64+ 11 68/67
55135 30/19 1218
43/42
82/52
61/60
<0.001 <0.001 0.234 0.785 0.724 0.210
133/84
9o/88
0.510
8/8 1/1 3/3 5.4 _+4.3
0.806
18/18
9/9
14/9 6/4 5/3 5.3 + 4.4
p Value
I 14/72 13/8 18/11 13/8
77/75 7/7 11/11
70/44
41/40
0.555
117/74 32/20
62161
0.034
0.936
7/7
26/25
9/6
14/14
47• 10 (n = 132)
45• 12 (n = 93)
0.273
LVEF, teil venlricular eieclion fraction.
hours (Table I). One hundred eighty-nine (73%) patients underwent angiography within 6 hours. Initial coronary angiography showed an occluded infarct artery in 211 (81%) patients and collateral circulation in 111 (44%) patients. The distribution of TIMI flow grade and the incidence of collateral circulation were not different between smokers and nonsmokers. When just 189 patients w h o under*vent angiography within 6 hours were analyzed, there was no difference in presence of a total occluded infarct artery between smokers and nonsmokers (80% vs 84%; p = 0.557). Eighty-one (31%) patients had multivessel disease. Muhivessel disease was less frequent in smokers than in nonsmokers (26% vs 39%; p = 0.035). Left ventriculograms adequate for analysis were obtained before initiation of thrombolytic therapy in 225 (87%) patients. Left ventricular ejection fraction was not different between smokers and nonsmokers.
Efficiency of thrombolytic therapy Despite that the distribution of TIMI flow grade was similar between smokers and nonsmokers, there was a tendency toward more frequent reperfusion (TIMI flow grade _>2) in smokers compared with nonsmokers after thrombolytic therapy (69~ vs 59~ p = 0.099) (Fig. 1). TIMI flow grade 3 was significantly more frequent in
American Heorl Journal Volume 134, Number 5, Port 1
smokers after thrombolytic therapy (32% vs 18%; p = 0.004). The average dose of urokinase used for thrombolytic therapy was smaller in smokers (61 + 30 x 104 IU vs 72 + 28 x 104 IU; p = 0.003). When just those 211 patients w h o had an occluded infarct artery on initial angiography were analyzed, TIblI flow grade 3 was still more frequent after thrombolytic therapy in smokers (24% vs 12%; p = 0.042). Adjunctive coronary angioplasty was performed in 68 (43%) patients among smokers and 53 (52%) patients among nonsmokers (p = 0.201). On final angiography, there was no difference in the incidence of reperfusion between smokers and nonsmokers (79~ vs 78%). However, TIMI flow grade 3 was still more frequent in smokers (56% vs 42%; p = 0.045).
In-hospital outcome Of 260 patients, 31 (12%) died during hospitalization. Fourteen patients died of p u m p failure, 11 of left ventricular wall rupture, four of reinfarction, and two of arrhythmia. Smokers were associated with lower in-hospital mortality rates compared with nonsmokers (8% vs 18%; p = 0.022). In-hospital deaths were also associated with older age (67 -+ 11 years vs 59 + 11 years; p < 0.001), female sex (35% vs 15%; p = 0.011), Killip class
957
958
Ishiharaelol.
Figure 2 100 Smokers 80
...."...........~........
Non-smokers
Cardiac survival
(%) 6O
p=O.O17 40
I
a
|
|
1
2
3
4
5
88% (130) 74% (73)
85% 020 710/. (70)
82% (114) 70% (62)
Years Smokers (n) Non-smokers (n)
90% (138) 79% (79)
90~ (137) "/5% (74)
Cardiac survival curve for smokers and nonsmokers. Long-term cardiac survival was significantly better in smokers than in nonsmokers.
>_2 (29% vs 12%; p = 0.026), multivessel disease (52O/ovs 28%; p = 0.016), lower left ventricular ejection fraction (41% + 10% vs 47% + 11%; p = 0.016), and final Tlbll flow grade -<2 compared with hospital survivors (81% vs 45%; p < 0.001). In-hospital deaths tended to be associated less with antecedent angina (39% vs 57%; p = 0.081). Reinfarction occurred in 13 (5%) patients during hospitalization. Reinfarction was less frequent in smokers than in nonsmokers, ahhough the difference was not significant (3% vs 8%; p = 0.161).
Long-term survival Follow-up was achieved for 153 (97%) smokers and 99 (97%) nonsmokers. One- and 5-year cardiac survival rates were 86% and 77%, respectively. Long-term cardiac survival was significantly better in smokers than in nonsmokers (Fig 2; p = 0.017). Univariate analysis showed that age <65 years (p < 0.001), Killip class 1 (p < 0.001), single-vessel disease (p = 0.011), left ventricular ejection fraction >-45% (p = 0.009), and final Tlbll flow grade 3 (p < 0.001) were associated with long-term cardiac survival. Multivariate analysis showed that age (p < 0.001) and final TIMI flow grade (p < 0.001) were independent predictors of long-term cardiac survival. Although the odds ratio continued to favor smokers, the comparison
was no longer significant (odds ratio 0.64 for smokers vs nonsmokers; 95% confidence interval 0.34 to 1.18; p = 0.150). When left ventricular ejection fraction was excluded from the analysis, age (p < 0.001), Killip class on admission (p =0.018), and final TlblI flow grade (p < 0.001) were independent predictors of long-term cardiac survival. Although it was not statistically significant, smoking tended to be associated with long-term survival (odds ratio 0.60; 95% confidence interval 0.34 to 1.05; p = 0.073).
Discussion Smoking is a well-established risk factor for the development of acute myocardial infarction. On the contrary, most studies have reported favorable outcome after acute myocardial infarction in smokers. This study strengthened the finding of previous studies. Although previous studies have reported favorable short-term outcome in smokers, findings of these studies were limited for those of
American HeartJoumol November 1997
Ishihara etal.
ous studies on short-term outcome to long-term outcome: smokers had a better long-term cardiac survival after acute myocardial infarction than had nonsmokers. Because smoking is associated with platelet aggregation and blood coagulability, coronary obstruction in smokers may be more thrombogenic and less atherosclerotic than in nonsmokers. 15,16 Indeed, angiographically recorded coronary artery disease was less extensive in smokers. Such a thrombogenic occlusion may tend to be reperfused spontaneously or therapeutically. Findings from previous studies in which a patent infarct artery was observed more frequently in smokers after thrombolytic therapy supported this hypothesis. However, these studies did not address coronary angiography before thrombolytic therapy and it remained unclear whether reperfusion was achieved spontaneously or therapeutically. In this stu~ty, we reviewed coronary angiograms both before and ~ifter thrombolytic therapy. On initial angiograms, the incidence of spontaneous reperfusion was not different between smokers and nonsmokers. After thrombolytic therapy, however, smokers were more likely tO have a reperfused infarct artery. We thus demonstrated that the infarct artery of smokers responded more efficiently to thrombolytic therapy compared with nonsmokers. Efficient response to thrombolytic therapy in smokers was more apparent when the incidence of TIMI grade 3 was compared. Recent studies have reported that only TlblI grade 3 but not TlbiI grade 2 was associated with favorable outcome after thrombolytic therapy in acute myocardial infarction37 These findings suggest that thrombolytic therapy is more effective in smokers. In turn, coronary angioplasty may be preferable to thrombolytic therapy in nonsmokers. 18 Smokers were not allowed to smoke cigarettes after the hospital admission. Sudden cessation of smoking may have had some favorable effects on vascular or hematologic systems. 15 Cessation of smoking could contribute to reducing the risk of reinfarction and death in smokers. 7,19 Smokers were associated with younger age, mate sex, and less multivessel disease compared with nonsmokers. These lower risk factors explain, at least in part, the favorable outcome in smokers. These findings, in turn, suggest that smoking may contribute to earlier acute myocardial infarction in healthier patients. It is important to note that, because smokers are more likely to suffer acute myocardial infarction, overall mortality rates from acute myocardial infarction are higher in smokers than in nonsmokers.
American Heart Journal Volume 134, Number 5, Parl I
Studylimitations This study is not a randomized but a retrospective study and suffers from the limitations of all nonrandomized, data-based analyses. The allocation of adjunctive coronary angioplasty was not randomized and was based on the physician's decision. We thus performed additional analyses. If only patients who underwent angioplasty were compared, 5-year survival rates were still more likely to be better in smokers than in nonsmokers (85% vs 72%; p = 0.082). Also, if only patients without angioplasty were compared, 5-year survival rates were still more likely to be better in smokers than in nonsmokers (80% vs 67%; p = 0.081). The dose of urokinase in each patient who underwent thrombolytic therapy Was not controlled; however, it is noteworthy that the average dose of urokinase used was smaller in smokers despite the fact that TIMI flow grade 3 was more frequent in smokers. It has been reported that smoking is a strong risk factor for the development of acute myocardial infarction, and cessation of smoking decreases the risk rapidly to that of nonsmokers. 2~ In turn, continuing smoking after acute myocardial infarction is associated with adverse outcome. 19 In this study, follow-up was concerned only with death and cause of death. Although all smokers were not allowed to smoke cigarettes during their hospitalization and were encouraged to quit, data about smoking status after hospital discharge were not obtained. Data about smoking status after hospital discharge may provide information about the impact of quitting smoking after acute myocardial infarction on long-term survival.
References 1. Kannel WB, McGee DL, Gastelli WP. I.atest perspective on cigarette smoking and cardiovascular disease: the Framingham Study. J Cardiac Rehabi11984;4:267-7Z 2. Doll R, Peto R. Mortality in relation to smoking: 20 years' observations of male British doctors. Br Med J 1976;2:1525-36. 3. AIMS Trial Study Group. Long4erm effects of iniravenous anistreplase in acute myocardial infarction: final reporl of the AIMS study. Lancet 1990;335:427-31. 4. Wilcox RG, Von Der [ippe G, Olsson CG, Jensen G, Skene AM, Hampton JR, for lhe Anglo-Scandinavian Study of Early Thrombolysis. Effects of alteplase in acute myocardial infarction: 6~nonth results from the ASSET study. Lancet 1990;335:1175-8. 5. Barbash GI, White HD, Modan M, Diaz R, Hampton JR, Heikkila J, et al., for the Investigation of the International Tissue Plasminogen Activator/Streptokinase Morbidily Trial. Significance of smoking in patients receiving lhrombolytic therapy for acute myocardial infarction: experience gleaned from the Investigation of the International Tissue Plasminogen Activator/Streptokinase Morbidily Trial. Circulation 1993;87:53-8.
959
960
Ishihara et al.
6. Molstad R Firstmyocardial in~ard~onin smokers.Eur Heart 1991;12:753-9. Z Mueller HS, Cohen LS, Braunwald E, Forman S, Feit F, RossA, el al., for the TIMI Investigators.Predictorsof early morbidily and mortalily after thrombofytic therapy for acute myocardial infarction: analysis of patient subgroup in lhe Thrombolysis in Myocardial InFarction(TIMI} Trial, Phase II. Circulation 1992;85:1254/o4. 8. Grines CL, Topol EJ, O'Neill WW, George BS, Kereiakes D, Phillips HR, et al. Effect of cigarette smoking on outcome after thrombolylic therapy for myocardial infarclion. Circulation 1995;91:298-303. 9. Barbash GI, ReinerJ, White HD, Wilcox RG, Armstrong PW, Sadowski Z, el al., for lhe GUSTO-I Investigators. Evaluation of paradoxic beneficial effects of smoking in patients receiving lhrombolytic therapy for acute myocardial infarclion: mechanismsof lhe "smoker's paradox" from lhe GUSTO-I Trial with angiographic insights.J Am Coll Cardio11995;26:1222-9. 10. Zahger D, Cercek B, Cannon CP, Jordan M, Davis V, Braunwald E, el oL, for the TIMI4 Investigators.How do smokers differ from nonsmokers in their response to lhrombolysis? (the TIMI4 Trial). Am J Cardiol
1995;75:232/o. 11. Goemz MJ, Karagounis IA, Allen A, Anderson JL, for the TEAM-2 Investigators. Effecl of cigarette smoking on coronary patency after lhromb61ytielherapy tar myocardial infarction. Am J Cardiol
1993;72:373-8. 12. De Chillou C, Rift P, Sadoul N, Elhevenot G, Feldmann L, Isaaz K, et al. Influence of cigarette smoking on rate of reopening of the infaradrelated coronary artery after myocardial infarclion: a multivariate analysiS.Am J Cardio11996;27:1662-8. 13. The TIMI Study Group. The Thrombolysis in Myocardial Infarction (TIMI) Trial. N Engl J Med 1985;312:932-6.
14. Sandier H, Dodge HT. The use o[ single plane angiocardiograms for the calculation of left ventricular volume in man. Am Heart J 1968;75:325-34. 15. Meade TW, Imeson J, Stirling Y. Effects of changes in smoking and other characleristics on dotting factors and the risk of ischemic heart disease. Lancet 1987;2:986-8. 16. FusterV, Chesebro JH, Frye RL, Elvebock LR. Platelet survival and the development of coronary arlery disease in.the young adult: effects of cigarette smoking, strong family history and medical therapy. Circulation 1981;63:546-51. 17 Karagounis L, Sorensen SG, Menlove R/, Moreno F, Anderson Jk Does Thrombolysis in Myocardial Infarction (TIMI) perfusion grade 2 represent a mostly patent artery or mostly occluded artery? Enzymatic and electrocardiographic evidence from the TEAM-2 Study. J Am Coil Cardio11992;19:1-10. 18. Bowers TR, Terrien EF,O'Neill WW0 Sachs D, Grines CL, for the PAMI Investigators. Effect of reperfusion modality on outcome in nonsmokers and smokerswith acute myocardial infarction (a PrimaryAng[oplasly in Myocardial Infarction [PAMI] substudy).Am J Cardio11996;78:511-5. 19. Rivers.IT,White HD, Cross DB, Williams BF, .Norris RM. Reinfarction after lhrombolylic therapy for acute myocardial infarction. ] Am Coil Cardio11990;16:340-8. 20. Rosenberg t, Kaufman DW, Helmrich SP, Shapiro S. The risk of myocardial infarction after quitting smoking in men under 55 years of age. N EnglJ Med 1985;313:1511-4. 21. Hermanson B, Omenn GS, Kronmal RA, Gersh BJ, br the participants in the Coronary Artery Surgery Study. Beneficial six-year outcome of smoking cessation in older men and women with coronary arlery disease: resultsfrom the CASS registry. N Engl J Med 1988;319:1365-9.
American Heart Journal November 1997
This study was undertaken to assesswhether reperfusion in smokers could be achieved spontaneously or therapeutically and to assesswhether favorable outcome in smokers could be sustained for years after infarction. We studied 260 patients with anterior myocardial infarction who underwent coronary angiography and thrombolysis within 24 hours after the onset of chest pain. There were 158 smokers and 102 nonsmokers. Smoking was associated more with men, younger age, and less multivessel disease. On initial angiography, the distribution of Thrombolysis in Myocardial Infarction grade was similar between smokers and nonsmokers. After thrombolysis, Thrombolysis in Myocardial Infarction grade 3 was more frequent in smokers (32% vs 18%; p = 0.004). In-hospital mortality rates were lower (8% vs 18%; p = 0.022) and long-term cardiac survival was better in smokers (5-year survival: 82% vs 70%; p = 0.022). Our data demonstrated that the infarct artery of smokers respgnded more efficiently to thrombotysis and favorable outcome in smokers was sustained throughout 5 years. (Am HeartJ 1997;134:955-60.)
It has been: well established that cigarette smoking is a risk factor for acute myocardial infarction and s u d d e n cardiac d e a t h ) ,2 Paradoxically, most studies have reported that, in patients w h o have had acute myocardial infarction, a smoking habit before acute myocardial infarction is a predictor of favorable short-term outcome after acute myocardial infarction. 3-1~ Recent studies, reviewing coronary angiography early after thrombolytic therapy, demonstrated that a patent infarct artery is more frequent in smokers and suggested m ' o possibilities: commonness of spontaneous reperfusion and efficient response of occlusive coronary thrombus to thrombolytic therapy. 8q2 Because of lack of angiograp h y before thrombolytic therapy, however, it remained unclear whether reperfusion in smokers was achieved spontaneously or therapeutically. To assess this issue, w e reviewed coronary angiograms that were performed early after acute myocardial infarction but before the initiation of thrombolytic therapy. In addition, although previous studies have demonstrated favorable shortterm outcome after acute myocardial infarction in smokers, it has not b e e n addressed whether favorable outcome in smokers is sustained for years. Because Fromthe Deportmentof Cardiobgy, HiroshimaCity Hospital Submitted Feb. 25, 1997; acceptedJune 17, 199Z Reprint requests:Masaharu Ishihara,MD, Departmentof Cardiology, HiroshimaCity Hospital 7-33, Moto-machi, Naka-ku, Hiroshima, 730 Japan. Copyright 9 1997 by Mosby-Year Book, Inc. 0002-8703/97/$5.00 + 0 411184850
smoking is a risk factor for acute myocardial infarction and sudden cardiac death, smokers may b e more likely to suffer cardiac death years after hospital discharge. We also assessed the effect of smoking on long-term prognosis in relation to acute angiographic findings.
Methods Study
patients
This study consisted of 260 consecutive patients with anterior wall acute myocardial infarction who underwent coronary angiography and thrombolytic therapy within 24 hours after the onset of chest pain between January 1981 and December 1990. Anterior wall acute myocardial infarction was diagnosed by chest pain consistent with ongoing myocardial ischemia persisting longer than 30 minutes, at least 1 mm ST segment elevation in at least m'o adjacent precordial electrocardiographic leads, and elevation of serum creatine kinase levels beyond twice the normal upper limit. Thrombolytic therapy performed after coronary angiography identified the left anterior descending artery as the infarct artery.
Cardiac catheterization Emergent cardiac catheterization was performed through the right femoral artery after heparin administration. Contrast left ventriculography was performed in the 30degree right anterior oblique projection. Selective coronary angiography was performed in multiple projections. Thrombolytic therapy was performed with urokinase. Usually, cumulative doses of urokinase were infused up to 96 x l0 t IU until reperfusion was achieved. Adjunctive coronary angioplasty was performed in 121 (47%) patients.
956
Ishihara et al.
Figure I 100
80
. . . . . . .
........
,
,.' 9176176176176 ..'.~ .o . . . o- .. o." ..~ 9 9 ~
.o ..- .." ~ . . " , o9 .~ .~ .- .. ~ ~176. ~ . . " . ,
".."~176176 .. . ...~ .~ .o . ~ ..~ .~1769176.., .- . . . ~ .- ..
' ..- ~176176176176 . . . . . ,.." ..'...'.~ . . . . .
9ITtMtO.- ."..".-".."..-'. .- . . . . . . . -
Incidence
6O
(%)
9 ........ o~ ~ ~ j 9" . . . ' 9 1 7 6 1 7 6
9
, . . . . . ~ ..~ . , ' . , ,, ..- ..- ,.~ , . " .
9-'...'..~149176176 ' . . , ' . . . ' . . . ~ 1 4 9
40
20
-ili-i~i-~i.~.~ :(1.2~.~.~i( 9 .. ~ ..... .o .o , . . . . . . . . . . .~ ~176.." . 9 . . . . , .
'~176149 .., .- .~ .~ ..... ~ ~ . . . - . . , ~176. " .
,~711@1,1N\ "- - - - " 9 : : ~ :
[:!:!:i:i:[:i:i:~[[!~:i:i:i:i:i:[:i:~:i
i
Smokers Non-smokers Before thrombolysis
Smokers Non-smokers After thrombolysis
TIMI flow grade before and after thrombolytic therapy. Left, Before thrombolytic therapy, distribution of TIMI flow grade was similar between smokers and nonsmokers (p = 0.936). Right, After thrombolytic therapy, there was tendency toward more frequent reperfusion in smokers compared with nonsmokers (69% vs 59%; p = 0.099). TIMI flow grade 3 was significantly more frequent in smokers after thrombolytic therapy (32% vs 18%; p = 0.004).
The allocation of thrombolysis or angioplasty was not randomized and was b a s e d on the physician's decision.
Angiographic analysis All coronary angiograms were reviewed by two angiographers without knowledge of the clinical variables. The peffusion status of the left anterior descending artery was determined in accordance with the Thrombolysis in Myocardial Infarction (TIMI) study classification. 13 Initial TIMI flow grade was assessed before the initiation of thrombolytic therapy. Efficiency of thrombolytic therapy was assessed on the angiogram after the final dose of urokinase was administered: 30 to 60 minutes after the initiation of urokinase. Final TIMI flow grade was assessed on the final shot of the angiography. An occluded artery was defined as TIMI flow grade <1. Reperfusion was defined as TIMI flow grade >_2. Multivessel coronary disease was defined as >_75% stenosis in one or more vessels remote from the infarct artery. Left main coronary disease was considered to be at least two-vessel involvement. Collateral circulation was considered to b e present w h e n there was partial or complete filling of the epicardial segment of the left anterior descending coronary artery. The 30-degree right anterior oblique end-diastolic and end-systolic left ventricular endocardial contours from sinus beats were traced by a single angiographer. The left ventricular ejection fraction was calculated acc9rding to the area-length method, t4
or clinical visit. Data for a full 5 years were available in most (233) patients. The primary end point was cardiac death defined as sudden death, death after acute myocardial infarction, or death after hospitalization for congestive heart failure9
Data analysis Statistical analysis was performed with the Z2 and t tests9 Kaplan-Meier estimates were used to construct a long-term su~,ival curve. Differences in long-term survival were assessed with the log-rank test. Cox proportional hazards regression was used to identify independent predictors of long-term survival. Differences were considered significant if p < 0.05. All group data are expressed as mean + SD.
Results Baseline clinical characteristics This study group included 260 patients. Of 260 patients, 158 (61%) had a smoking habit at the time of acute myocardial infarction (smokers) and 102 (39~ did not (nonsmokers) (Table I). Average tobacco consumption was 23 +_ 15 cigarettes per day in smokers. Smokers were more likely to be male and younger compared with nonsmokers. There was no difference in hypertension, diabetes, previous infarction, antecedent angina, or Killip class on admission to the study.
Follow-up
Initial angiographic findings
Follow-up was conducted between January 1995 a n d April 1995. Follow-up was achieved for 252 (97%) patients by mail
Average time from the onset of acute myocardial infarction to emergency angiography was 5.3 -+ 4.4
American Heart Journd November 1997
Ishihara et al.
Table h Baseline clinical and initial ongiographic findings
Age (yr) Men {n/%) Hypertension (n/%) Diabetes (n/%) Previous infarction (n/%) Antecedent angina (n/%) Killlp class (n/%) 1 2 3 4 Time to angiography (hr) Initial TIMI flow grade 0 I 2 3 I Collateral circulation (n/%) No. diseased Vessels {n/%) 1 2 3 LVEF (%}
Smokers (n = 158)
Nonsmokers (n = 102)
57+ 11 146/92
64+ 11 68/67
55135 30/19 1218
43/42
82/52
61/60
<0.001 <0.001 0.234 0.785 0.724 0.210
133/84
9o/88
0.510
8/8 1/1 3/3 5.4 _+4.3
0.806
18/18
9/9
14/9 6/4 5/3 5.3 + 4.4
p Value
I 14/72 13/8 18/11 13/8
77/75 7/7 11/11
70/44
41/40
0.555
117/74 32/20
62161
0.034
0.936
7/7
26/25
9/6
14/14
47• 10 (n = 132)
45• 12 (n = 93)
0.273
LVEF, teil venlricular eieclion fraction.
hours (Table I). One hundred eighty-nine (73%) patients underwent angiography within 6 hours. Initial coronary angiography showed an occluded infarct artery in 211 (81%) patients and collateral circulation in 111 (44%) patients. The distribution of TIMI flow grade and the incidence of collateral circulation were not different between smokers and nonsmokers. When just 189 patients w h o under*vent angiography within 6 hours were analyzed, there was no difference in presence of a total occluded infarct artery between smokers and nonsmokers (80% vs 84%; p = 0.557). Eighty-one (31%) patients had multivessel disease. Muhivessel disease was less frequent in smokers than in nonsmokers (26% vs 39%; p = 0.035). Left ventriculograms adequate for analysis were obtained before initiation of thrombolytic therapy in 225 (87%) patients. Left ventricular ejection fraction was not different between smokers and nonsmokers.
Efficiency of thrombolytic therapy Despite that the distribution of TIMI flow grade was similar between smokers and nonsmokers, there was a tendency toward more frequent reperfusion (TIMI flow grade _>2) in smokers compared with nonsmokers after thrombolytic therapy (69~ vs 59~ p = 0.099) (Fig. 1). TIMI flow grade 3 was significantly more frequent in
American Heorl Journal Volume 134, Number 5, Port 1
smokers after thrombolytic therapy (32% vs 18%; p = 0.004). The average dose of urokinase used for thrombolytic therapy was smaller in smokers (61 + 30 x 104 IU vs 72 + 28 x 104 IU; p = 0.003). When just those 211 patients w h o had an occluded infarct artery on initial angiography were analyzed, TIblI flow grade 3 was still more frequent after thrombolytic therapy in smokers (24% vs 12%; p = 0.042). Adjunctive coronary angioplasty was performed in 68 (43%) patients among smokers and 53 (52%) patients among nonsmokers (p = 0.201). On final angiography, there was no difference in the incidence of reperfusion between smokers and nonsmokers (79~ vs 78%). However, TIMI flow grade 3 was still more frequent in smokers (56% vs 42%; p = 0.045).
In-hospital outcome Of 260 patients, 31 (12%) died during hospitalization. Fourteen patients died of p u m p failure, 11 of left ventricular wall rupture, four of reinfarction, and two of arrhythmia. Smokers were associated with lower in-hospital mortality rates compared with nonsmokers (8% vs 18%; p = 0.022). In-hospital deaths were also associated with older age (67 -+ 11 years vs 59 + 11 years; p < 0.001), female sex (35% vs 15%; p = 0.011), Killip class
957
958
Ishiharaelol.
Figure 2 100 Smokers 80
...."...........~........
Non-smokers
Cardiac survival
(%) 6O
p=O.O17 40
I
a
|
|
1
2
3
4
5
88% (130) 74% (73)
85% 020 710/. (70)
82% (114) 70% (62)
Years Smokers (n) Non-smokers (n)
90% (138) 79% (79)
90~ (137) "/5% (74)
Cardiac survival curve for smokers and nonsmokers. Long-term cardiac survival was significantly better in smokers than in nonsmokers.
>_2 (29% vs 12%; p = 0.026), multivessel disease (52O/ovs 28%; p = 0.016), lower left ventricular ejection fraction (41% + 10% vs 47% + 11%; p = 0.016), and final Tlbll flow grade -<2 compared with hospital survivors (81% vs 45%; p < 0.001). In-hospital deaths tended to be associated less with antecedent angina (39% vs 57%; p = 0.081). Reinfarction occurred in 13 (5%) patients during hospitalization. Reinfarction was less frequent in smokers than in nonsmokers, ahhough the difference was not significant (3% vs 8%; p = 0.161).
Long-term survival Follow-up was achieved for 153 (97%) smokers and 99 (97%) nonsmokers. One- and 5-year cardiac survival rates were 86% and 77%, respectively. Long-term cardiac survival was significantly better in smokers than in nonsmokers (Fig 2; p = 0.017). Univariate analysis showed that age <65 years (p < 0.001), Killip class 1 (p < 0.001), single-vessel disease (p = 0.011), left ventricular ejection fraction >-45% (p = 0.009), and final Tlbll flow grade 3 (p < 0.001) were associated with long-term cardiac survival. Multivariate analysis showed that age (p < 0.001) and final TIMI flow grade (p < 0.001) were independent predictors of long-term cardiac survival. Although the odds ratio continued to favor smokers, the comparison
was no longer significant (odds ratio 0.64 for smokers vs nonsmokers; 95% confidence interval 0.34 to 1.18; p = 0.150). When left ventricular ejection fraction was excluded from the analysis, age (p < 0.001), Killip class on admission (p =0.018), and final TlblI flow grade (p < 0.001) were independent predictors of long-term cardiac survival. Although it was not statistically significant, smoking tended to be associated with long-term survival (odds ratio 0.60; 95% confidence interval 0.34 to 1.05; p = 0.073).
Discussion Smoking is a well-established risk factor for the development of acute myocardial infarction. On the contrary, most studies have reported favorable outcome after acute myocardial infarction in smokers. This study strengthened the finding of previous studies. Although previous studies have reported favorable short-term outcome in smokers, findings of these studies were limited for those of
American HeartJoumol November 1997
Ishihara etal.
ous studies on short-term outcome to long-term outcome: smokers had a better long-term cardiac survival after acute myocardial infarction than had nonsmokers. Because smoking is associated with platelet aggregation and blood coagulability, coronary obstruction in smokers may be more thrombogenic and less atherosclerotic than in nonsmokers. 15,16 Indeed, angiographically recorded coronary artery disease was less extensive in smokers. Such a thrombogenic occlusion may tend to be reperfused spontaneously or therapeutically. Findings from previous studies in which a patent infarct artery was observed more frequently in smokers after thrombolytic therapy supported this hypothesis. However, these studies did not address coronary angiography before thrombolytic therapy and it remained unclear whether reperfusion was achieved spontaneously or therapeutically. In this stu~ty, we reviewed coronary angiograms both before and ~ifter thrombolytic therapy. On initial angiograms, the incidence of spontaneous reperfusion was not different between smokers and nonsmokers. After thrombolytic therapy, however, smokers were more likely tO have a reperfused infarct artery. We thus demonstrated that the infarct artery of smokers responded more efficiently to thrombolytic therapy compared with nonsmokers. Efficient response to thrombolytic therapy in smokers was more apparent when the incidence of TIMI grade 3 was compared. Recent studies have reported that only TlblI grade 3 but not TlbiI grade 2 was associated with favorable outcome after thrombolytic therapy in acute myocardial infarction37 These findings suggest that thrombolytic therapy is more effective in smokers. In turn, coronary angioplasty may be preferable to thrombolytic therapy in nonsmokers. 18 Smokers were not allowed to smoke cigarettes after the hospital admission. Sudden cessation of smoking may have had some favorable effects on vascular or hematologic systems. 15 Cessation of smoking could contribute to reducing the risk of reinfarction and death in smokers. 7,19 Smokers were associated with younger age, mate sex, and less multivessel disease compared with nonsmokers. These lower risk factors explain, at least in part, the favorable outcome in smokers. These findings, in turn, suggest that smoking may contribute to earlier acute myocardial infarction in healthier patients. It is important to note that, because smokers are more likely to suffer acute myocardial infarction, overall mortality rates from acute myocardial infarction are higher in smokers than in nonsmokers.
American Heart Journal Volume 134, Number 5, Parl I
Studylimitations This study is not a randomized but a retrospective study and suffers from the limitations of all nonrandomized, data-based analyses. The allocation of adjunctive coronary angioplasty was not randomized and was based on the physician's decision. We thus performed additional analyses. If only patients who underwent angioplasty were compared, 5-year survival rates were still more likely to be better in smokers than in nonsmokers (85% vs 72%; p = 0.082). Also, if only patients without angioplasty were compared, 5-year survival rates were still more likely to be better in smokers than in nonsmokers (80% vs 67%; p = 0.081). The dose of urokinase in each patient who underwent thrombolytic therapy Was not controlled; however, it is noteworthy that the average dose of urokinase used was smaller in smokers despite the fact that TIMI flow grade 3 was more frequent in smokers. It has been reported that smoking is a strong risk factor for the development of acute myocardial infarction, and cessation of smoking decreases the risk rapidly to that of nonsmokers. 2~ In turn, continuing smoking after acute myocardial infarction is associated with adverse outcome. 19 In this study, follow-up was concerned only with death and cause of death. Although all smokers were not allowed to smoke cigarettes during their hospitalization and were encouraged to quit, data about smoking status after hospital discharge were not obtained. Data about smoking status after hospital discharge may provide information about the impact of quitting smoking after acute myocardial infarction on long-term survival.
References 1. Kannel WB, McGee DL, Gastelli WP. I.atest perspective on cigarette smoking and cardiovascular disease: the Framingham Study. J Cardiac Rehabi11984;4:267-7Z 2. Doll R, Peto R. Mortality in relation to smoking: 20 years' observations of male British doctors. Br Med J 1976;2:1525-36. 3. AIMS Trial Study Group. Long4erm effects of iniravenous anistreplase in acute myocardial infarction: final reporl of the AIMS study. Lancet 1990;335:427-31. 4. Wilcox RG, Von Der [ippe G, Olsson CG, Jensen G, Skene AM, Hampton JR, for lhe Anglo-Scandinavian Study of Early Thrombolysis. Effects of alteplase in acute myocardial infarction: 6~nonth results from the ASSET study. Lancet 1990;335:1175-8. 5. Barbash GI, White HD, Modan M, Diaz R, Hampton JR, Heikkila J, et al., for the Investigation of the International Tissue Plasminogen Activator/Streptokinase Morbidily Trial. Significance of smoking in patients receiving lhrombolytic therapy for acute myocardial infarction: experience gleaned from the Investigation of the International Tissue Plasminogen Activator/Streptokinase Morbidily Trial. Circulation 1993;87:53-8.
959
960
Ishihara et al.
6. Molstad R Firstmyocardial in~ard~onin smokers.Eur Heart 1991;12:753-9. Z Mueller HS, Cohen LS, Braunwald E, Forman S, Feit F, RossA, el al., for the TIMI Investigators.Predictorsof early morbidily and mortalily after thrombofytic therapy for acute myocardial infarction: analysis of patient subgroup in lhe Thrombolysis in Myocardial InFarction(TIMI} Trial, Phase II. Circulation 1992;85:1254/o4. 8. Grines CL, Topol EJ, O'Neill WW, George BS, Kereiakes D, Phillips HR, et al. Effect of cigarette smoking on outcome after thrombolylic therapy for myocardial infarclion. Circulation 1995;91:298-303. 9. Barbash GI, ReinerJ, White HD, Wilcox RG, Armstrong PW, Sadowski Z, el al., for lhe GUSTO-I Investigators. Evaluation of paradoxic beneficial effects of smoking in patients receiving lhrombolytic therapy for acute myocardial infarclion: mechanismsof lhe "smoker's paradox" from lhe GUSTO-I Trial with angiographic insights.J Am Coll Cardio11995;26:1222-9. 10. Zahger D, Cercek B, Cannon CP, Jordan M, Davis V, Braunwald E, el oL, for the TIMI4 Investigators.How do smokers differ from nonsmokers in their response to lhrombolysis? (the TIMI4 Trial). Am J Cardiol
1995;75:232/o. 11. Goemz MJ, Karagounis IA, Allen A, Anderson JL, for the TEAM-2 Investigators. Effecl of cigarette smoking on coronary patency after lhromb61ytielherapy tar myocardial infarction. Am J Cardiol
1993;72:373-8. 12. De Chillou C, Rift P, Sadoul N, Elhevenot G, Feldmann L, Isaaz K, et al. Influence of cigarette smoking on rate of reopening of the infaradrelated coronary artery after myocardial infarclion: a multivariate analysiS.Am J Cardio11996;27:1662-8. 13. The TIMI Study Group. The Thrombolysis in Myocardial Infarction (TIMI) Trial. N Engl J Med 1985;312:932-6.
14. Sandier H, Dodge HT. The use o[ single plane angiocardiograms for the calculation of left ventricular volume in man. Am Heart J 1968;75:325-34. 15. Meade TW, Imeson J, Stirling Y. Effects of changes in smoking and other characleristics on dotting factors and the risk of ischemic heart disease. Lancet 1987;2:986-8. 16. FusterV, Chesebro JH, Frye RL, Elvebock LR. Platelet survival and the development of coronary arlery disease in.the young adult: effects of cigarette smoking, strong family history and medical therapy. Circulation 1981;63:546-51. 17 Karagounis L, Sorensen SG, Menlove R/, Moreno F, Anderson Jk Does Thrombolysis in Myocardial Infarction (TIMI) perfusion grade 2 represent a mostly patent artery or mostly occluded artery? Enzymatic and electrocardiographic evidence from the TEAM-2 Study. J Am Coil Cardio11992;19:1-10. 18. Bowers TR, Terrien EF,O'Neill WW0 Sachs D, Grines CL, for the PAMI Investigators. Effect of reperfusion modality on outcome in nonsmokers and smokerswith acute myocardial infarction (a PrimaryAng[oplasly in Myocardial Infarction [PAMI] substudy).Am J Cardio11996;78:511-5. 19. Rivers.IT,White HD, Cross DB, Williams BF, .Norris RM. Reinfarction after lhrombolylic therapy for acute myocardial infarction. ] Am Coil Cardio11990;16:340-8. 20. Rosenberg t, Kaufman DW, Helmrich SP, Shapiro S. The risk of myocardial infarction after quitting smoking in men under 55 years of age. N EnglJ Med 1985;313:1511-4. 21. Hermanson B, Omenn GS, Kronmal RA, Gersh BJ, br the participants in the Coronary Artery Surgery Study. Beneficial six-year outcome of smoking cessation in older men and women with coronary arlery disease: resultsfrom the CASS registry. N Engl J Med 1988;319:1365-9.
American Heart Journal November 1997