Geriatric patients with acute myocardial infarction: Cardiac risk factor profiles, presentation, thrombolysis, coronary interventions, and prognosis

Sumita D. Paul, MD, MPH, Patrick T. O'Gara, MD, Zakwan A. Mahjoub, MD, Thomas G. DiSalvo, MD,

Christopher J. O'Donnell, MD, MPH, John B. Newell, AB, Gerardo Villarreal-Levy, MD,

A.J. Conrad Smith, MD, Nicholas I. Kondo, MD, Marta Cararach, MD, Laura Ferrer, MD, and

Kim A. Eagle, MD Boston, Mass., and Ann Arbor, Mich.

Elderly patients have a higher mortality after acute myocar- dial infarction (MI) yet are treated less aggressively than younger patients. To determine (1) the risk-factor profiles, (2) presentation, (3) management, and (4) hospital outcomes for the elderly (->75 years) compared with middle aged (66 to 74 years) and younger (-<65 years) patients in the 1990s, we studied 561 consecutive patients with acute MI. Compared with younger patients, the elderly more frequently had con- gestive heart failure (40% vs 14%; p < 0.00001) and non-Q wave infarctions (76% vs 56%; p < 0.005), received throm- bolysis (9% vs 34%; p < 0.0001), and underwent catheteriza- tion (35% vs 73%; p < 0.00001), percutaneous transluminal coronary angioplasty (9% vs 31%; p < 0.0002), and coronary artery bypass grafting (5% vs 15%; p < 0.03) less frequently. Those who did not receive thrombolysis all had contraindi- cations. Mortality was higher in the elderly (19% vs 5%; p < 0.004), especially among those who did not receive thrombolysis (20% vs 7%; p < 0.03). Multivariate predictors of mortality included age, and congestive heart failure, in addition, when clinical course and management variables were considered, use of the intraaortic balloon pump was a predictor of mortality, whereas undergoing coronary an- giography was a negative predictor (relative risk, 0.3; 95% confidence intervals, 0.1 to 0.6). (AM HEART J 1996;131:710- 5.)

In the United States alone, 675,000 patients are hospitalized with acute myocardial infarction (MI) each year. 1 In addition, 55% of all patients with an MI are ->65 years old. 2 Although prior studies have found that the elderly are less likely to receive thrombolytics or coronary angiography after an MI,3, 4 the specific reasons for the less aggressive care

From the Cardiac Unit, Department of Medicine, Massachusetts General Hospital, and the Division of Cardiology, Department of Medicine, Univer- sity of Michigan Medical Center,

Received for publication Jan. 4, 1995; accepted June 1, 1995. Reprint requests: Sumita D. Paul, MD, MPH, Cardiac Unit, Ellison 908, Massachusetts General Hospital, Boston, MA 02114.

Copyright © 1996 by Mosby-Year Book, Inc. 0002-8703/96/$5.00 + 0 4/1/67902

are not well defined. It is not clear if physicians have a bias against using thrombolytics in the elderly or if therapeutic decisions consider the presence of valid contraindications to thrombolysis and not advanced age alone. Physicians may have insufficient data for management decisions about patients older than 75 years.

Furthermore, with the aging population in the United States, it is prudent to examine the risk-fac- tor profiles and clinical presentation of elderly pa- tients to develop appropriate prevention strategies for this group at high risk. Such information may be valuable for targeting public health programs aimed at risk-factor modification to specific age groups. The aims of this study were to determine (1) cardiac risk-factor profiles, (2) presentation, (3) manage- ment strategies, and (4) hospital outcomes in the elderly (>-75 years) compared with middle aged (66 to 74 years) and younger (---65 years) patients with acute MI who were admitted to a large university teaching hospital.

METHODS Study population. We studied 561 patients with acute

MI who were identified prospectively by daily rounds and enrolled in a Myocardial Infarction Registry at Massachu- setts General Hospital between 1991 and 1992 after informed consent. The study was approved by the Sub- committee on Human Studies. There were 248 patients who were -<65 years of age ("young" subset), 169 were be- tween 66 and 74 years old ("middle-aged" subset), and 144 patients were >-75 years ("elderly" subset). MI was defined by an elevation of the creatine kinase-MB index to >-3%, in addition to (1) a history compatible with MI or (2) electro- cardiographic abnormalities or both. The abnormalities are defined as (1) evolution of pathologic Q waves (->0.04 second), or (2) ->0.1 mV ST-segment elevation in contigu- ous leads, or (3) >-0.1 mV ST-segment depression or defi- nite T-wave inversion or both. Electrocardiographic crite-

710

Volume 131, Number 4 American Heart Journal Paul et al. 711

ria for thrombolysis were defined as ->0.1 mV ST-segment elevation in two contiguous leads or new (or presumed new) left bundle branch block. Thrombolytic treatment was routinely given by the emergency room physicians for up to 6 hours after the onset of symptoms during the 18 months of this study (current indications differ).

Data collection and statistical analysis. Data for his- tory, hospital course, and outcome were obtained by chart review by investigators who were not aware of the hypothe- ses being tested. Data were obtained using special data- collection forms and maintained in the Myocardial Infarc- tion Registry. Data quality control was maintained by involving trained physicians in the collection of clinical data. These physicians had undergone testing for inter- and intrareviewer reliability by using samples of patients (requiring >95% reliability rates). Reliability checks were also done for data entry.

Statistical analysis was done using the BMDP statisti- cal program (Los Angeles, Calif.). 5 A stratified analysis was performed from the source of the patient (emergency department vs hospital transfer). Fisher's exact tests were used to compare categorical variables. Student's t test was used to compare independent means of continuous vari- ables. The p values reflect two-sided tests. Stepwise mul- tiple logistic regression was done to determine indepen- dent predictors of hospital mortality (p < 0.05), by using the BMDP statistical package (program LR). Univariate predictors of mortality with a p < 0.1 were allowed to en- ter the model.

RESULTS Clinical features. The differences in clinical fea tures

are given in Tables I and II. The elderly were more likely to have a h is tory of congestive hea r t failure. However , an infarct ion in the elderly pa t ien t was significantly less l ikely to be associated with the car- diac r isk factors of a family his tory of MI, hypercho- lesterolemia, or smoking. In addition, as shown in Fig. 1, there was an increase in the proport ion ofpa- t ients wi th congestive hea r t fai lure and a decrease in the proport ion wi th chest pa in wi th increasing age. The elderly were also much more likely to evolve n o n - Q wave infarcts (Fig. 1). There were no signif- icant age differences in the proport ion of pa t ien ts with a prior MI, diabetes mellitus, or hyper tension.

Management. Sixty-four percent or 50 of the 78 eld- erly pa t ien ts seen in the emergency depa r tmen t did not mee t e lectrocardiographic (ECG) cri ter ia for thrombolysis; even among the 28 pa t ien ts who me t the criteria, 7 pa t ien ts (9% of all elderly) received it. This was in cont ras t to a 44% (n = 47) use of lytic t h e r a p y among young pa t ien ts f rom the same cohort who met ECG cri ter ia (n = 108; 34% of all young pa- t ients received i t ;p < 0.00003, as shown in Fig. 1). All the elderly pa t ien ts who me t ECG cri ter ia for throm- bolysis bu t did not receive it had contraindicat ions as

M k ~ 8O0 (SS-74 yrs) ~___ B d e ~ (>75 yr=)

(%) *p<=oo

0

0 CHF

0

34% 21%

Fig. 1. Comparisons of young (-<65 years) versus middle- aged (66 to 74 years) versus elderly (->75 years) patients with respect to presentation, Q-wave infarcts, and use of thrombolysis (p values represent overall comparison and comparison of elderly vs young).

perceived by the emergency d ep a r tm en t physician (Table III). Elder ly pa t ien ts who did not receive thrombolysis had a much h igher mor ta l i ty t h a n younger pat ients who did not receive lyric t h e r a p y (20% vs 7%; p < 0.04; Table IV).

In addition, among pat ients admi t ted from the emergency depar tment , 102 (73%) pa t ien ts ---65 years u n d e rw en t coronary angiography versus 27 (35%) of the elderly (p < 0.00001; Fig. 2). This oc-

April 1996 712 Paul 8t at. American Heart Journal

YOUNG MIDDLE AGE ELDERLY (n=139) (n=75) (n=78)

27% 33% 5%

I2% 3% 28% 14% 1% 22% 9% 65%

No Coronary Angiography (p<o.ooool]

Coronary Angiography+ PTCA alone (p <a oo-/)

~ ] Coronary Angiography+CABG alone (p < o.o5)

Coronary Angiography+PTCA & CABG

Fig. 2. Comparison of young (-<65 years) versus middle-aged (66 to 74 years) versus elderly (->75 years) patients with respect to coronary intervention (p values represent overall comparison and comparison of elderly vs young).

Table I. Clinical features of patients seen in the emergency department (n = 292)

Y oung Middle-aged Elder ly (n = 139) (n = 75) (n = 78) p Value

Women 30 (22%) 21 (28%) 28 (36%) <0.08 History of MI 43 (31%) 32 (43%) 33 (42%) NS History of CHF 14 (10%) 12 (16%) 25 (32%) <0.0003 Family history 59 (42%) 25 (33%) 17 (22%) <0.008 Hyperlipidemia 71 (51%) 20 (27%) 17 (22%) <0.00001 Smoking 86 (62%) 38 (51%) 33 (42%) <0.009 Diabetes me]]itus 30 (22%) 25 (33%) 18 (23%) NS Hypertension 75 (54%) 39 (52%) 50 (64%) NS

Young, -<65 years; middle-aged, 66 to 74 years; elderly, ->75 years; CHF, congestive heart failure; NS, not significant.

curred in spite of the fact that elderly patients had a greater relative proportion of non-Q wave infarcts. As shown in Fig. 2, younger patients were three times as likely to undergo percutaneous translumi- nal coronary angioplasty (PTCA) or coronary artery bypass graft (CABG).

There were significant differences in mortality of patients who underwent cardiac catheterization, PTCA, or CABG in the elderly compared with the young subset (Table IV). The only young patient who died after PTCA had cardiogenic shock. Of the four elderly patients who died after PTCA, two had car- diogenic shock, the third had a right ventricular in- farction with biventricular failure, complete heart block, and unsuccessful right coronary artery PTCA, and the fourth patient developed progressive renal failure after redo CABG.

Hospital outcomes. An increasing gradient of mor- talitY was seen with increasing age (5% in the young vs 8% in the middle aged and 19% in the elderly [p < 0.001]) for patients admitted from the emer- gency department. However, the incidence of recur- rent angina was less frequent among the elderly, (31% vs 49% in the middle aged and 44% in the young [p < 0.05]); even though the elderly had a propor- tionately greater number of non-Q wave infarctions. This may partly explain the lesser use of PTCA in this group. Table V displays the results of the univariate analysis for predictors of hospital mortal- ity. Multiple logistic regression identified increasing age, congestive heart failure, and the use ofintraaor- tic balloon pump to be independent predictors of hospital mortality (Table VI). Coronary angiography was a negative predictor of mortality, with a relative

Volume 131, Number 4 American Heart Journal PGul et al. 713

Table II. Clinical features of patients who were transferred from other hospitals (n = 269)

Y oung Middle-aged Elderly (n = 109) (n = 94) (n = 66) p Value

Women 28 (26%) 36 (38%) 35 (53%) <0.0007 History of MI 35 (32%) 32 (34%) 29 (44%) NS History of CHF 12 (11%) 17 (18%) 17 (26%) <0.039 Family history 59 (54%) 36 (38%) 20 (30%) <0.0047 Hyperlipidemia 66 (61%) 38 (40%) 19 (29%) <0.00007 Smoking 68 (62%) 50 (53%) 17 (26%) <0.00001 Diabetes mellitus 21 (19%) 29 (31%) 16 (24%) NS Hypertension 60 (55%) 66 (70%) 39 (59%) <0.034

Young, -<65 years; middle-aged, 66 to 74 years; elderly, ->75 years; CHF, congestive heart failure; NS, not significant.

risk of 0.3 (95% confidence intervals: 0.1 to 0.6). When variables of the clinical presentation alone were allowed to enter the multivariate model, age and congestive hear t failure were the only indepen- dent predictors of mortality.

DISCUSSION Presentation. We found that the elderly are signif-

icantly less likely to report typical chest pain, have a substantially higher incidence of congestive hear t failure, and evolve a greater proportion of non-Q wave infarctions. These clinical features undoubt- edly affect the type of care provided and also affect the subsequent prognosis.

Our study adds to previous work by demonstrating an age difference in the relative frequencies of cardiac risk factors. Whereas family history, hyper- lipidemia, and smoking may be more important among younger patients, hypertension is more fre- quent among the middle-aged and elderly patients with acute MI. In addition, a greater proportion of the middle-aged and elderly patients were diabetic, although this did not reach statistical significance. A better understanding of the prevalence of various cardiac risk factors among patients with acute MI in a large university teaching hospital may help to de- velop secondary prevention programs to target dif- ferent age groups.

Management. A tendency toward a less aggressive management approach in the elderly (or more ag- gressive in the young) was observed in our study. The large-scale impact of such a difference on subsequent prognosis is unknown. However, large multicenter studies, particularly the Second International Study of Infarct Survival (ISIS-2), 6 have clearly established mortality reduction with thrombolysis in elderly pa- tients after MI. The overall 5-week mortality for pa- tients ---70 years of age, who received both streptoki- nase and aspirin, was 15.8% compared with 23.8% in the placebo group (p < 0.001).

Table IlL Contraindications to the use of thrombolytic therapy in eligible elderly patients admitted from the emergency department

Number of patients (%)* Contraindication

1 (5) Bleeding disorder 3 (14) Hypotension 5 (24) Age alone 4 (19) Occult blood-positive stools 8 (38) Duration of symptoms >6 hours

TOTAL 21 (75)

*Reflects the percentage of the total number (N = 28) of elderly patients seen in the emergency department and meeting ECG criteria for thrombo- lysis.

The key issue for the elderly is that only a small fraction have symptoms and ECG findings that qualify them for thrombolytic therapy. This m a y be because of a reduced incidence of sudden plaque rupture 7 or because extensive collaterals protect against t ransmural infarction. Although atypical presentation of elderly patients can explain many of these differences, there may remain instances in which acute reperfusion therapies are withheld be- cause of unsubstantiated concerns about the high risk of side effects. Among 21 elderly patients thought to have contraindications to thrombolysis, age alone was the only contraindication in 5 (24%), and in 8 (38%), duration of symptoms >6 hours was a con- traindication (Table III). Current knowledge sug- gests that such patients are likely candidates for thrombolytic therapy.

Other questions arise with the disparate use ofin- vasive procedures in the elderly. Are we performing too few procedures in the elderly or too many in the young? Coronary angiography was performed in 35% of patients --75 years and in 73% of patients <65 years (p < 0.00001). In addition, younger patients were much more likely to undergo PTCA or CABG

April 1996 714 P a u l et al. American Heart Journal

Table IV. Mortality rate after hospital interventions for patients admitted from the emergency department

Mortality

Young Middle-aged Elderly Intervention (n = 139) (n = 75) (n = 78) p Value

Thrombolysis 1/47 (2%) 0/16 (0%) 1/7 (14%) NS No thrombolysis 6/92 (7%) 6/59 (10%) 14/71 (20%) <0.033 Catheterization 3/102 (3%) 3/50 (6%) 4/27 (15%) <0.048 PTCA 1/43 (2%) 1/17 (6%) 2/7 (29%) <0.035 CABG 0/21 (0%) 1/11 (9%) 2/4 (50%) <0.009

Young, -<65 years; middle-aged, 66 to 74 years; elderly, ->75 years; NS, not significant.

Table V. Univariate predictors of mortality

D i e d S u r v i v e d

n = 53 (%) n = 508 (%) p V a l u e

Age >-75 years 25 (47) 119 (23) <0.0004 Chest pain 30 (57) 379 (75) <0.0086 CHF on presentat ion 22 (42) 91 (18) <0.0002 Prior MI 24 (45) 180 (35) NS Prior CABG 3 (6) 76 (15) <0.06 Prior PTCA 4 (8) 37 (7) NS Q wave on ECG 22 (42) 179 (35) NS Thrombolysis 9 (17) 142 (28) <0.1 Recurrent angina 19 (36) 271 (53) <0.02 IABP 19 (36) 89 (18) <0.0028 Coronary angiography 29 (55) 392 (77) <0.0007 PTCA 10 (19) 163 (32) <0.059 CABG 9 (17) 82 (16) NS

CHF, Congestive heart failure; IABP, intraaortic balloon pump; NS, not significant.

compared with the elderly. Although elderly patients (->75 years) in the MITI Project s also underwent cor- onary angiography less frequently than did younger patients (<55 years; 22% vs 76%; p < 0.0001), the reasons for this differential were not evident from their study. However, we have been able to demon- strate a significantly lower incidence of postinfarc- tion ischemia in elderly patients compared wi th younger patients. This may help to explain the less aggressive use of coronary angiography among the elderly.

Prognosis. Coronary heart disease is the leading cause of death among the elderly. 9 Elderly patients at our institution had a nearly fourfold greater mor- tality than did younger patients. These results are similar to those of Smith et al., 1° who compared the clinical course and outcomes at 1 year for patients older than 75 years versus those between 65 and 75 years old. They found a 20% hospital mortality for those older than 75 years versus a 12% mortality for those between 65 and 75 years old. Pashos et al. 11

Table VI. Multivariate predictor s of mortality

95% Standard Relative Confidence

Coefficient error p Value risk intervals

Age* 0.05 0.01 <0.0008 - - - - CHF 0.8 0.3 <0.01 2.3 1.2-4.2 IABP 1.7 0.4 <0.0001 5.4 2.5-12 CATH -1.2 0.4 <0.0002 0.3 0.1-0.6

CATH, Coronary angiography; CHF, congestive heart failure; IABP, intraaortic balloon pump. *Age as a continuous variable.

found a 30-day mortality of 23% in their 1990 study of elderly patients insured by Medicare.

Although a higher mortality after an acute MI has been documented for the elderly, the mechanism for this excess risk is still not certain. However, several physiologic features of aging may predispose to the development of congestive heart failure, arrhyth- mias, and recurrent ischemia, 3 and thus relate to this excess risk. These changes include increases in left ventricular mass, 12-14 reduced diastolic compli- ance,15, 16 and increased peripheral vascular resis- tance, 17 all of which increase cardiac work. Aging is associated also with a decline in heart rate variabil- ity,lS, 19 which is a marker of increased susceptibility to sudden death 19' 20 and all-cause mortality after MI. 21 Reduced vagal tone in the elderly has also been shown to predispose to a higher incidence of sudden death after MI. 22 Future randomized trials should include the elderly so that management strategies that help to reduce the high mortality among the elderly may be developed.

Study limitations. Because patients in this study were enrolled from a University teaching hospital, the results obtained may not be generalizable to all other hospitals or settings. Furthermore, it is not possible to infer a causal relation between older age and worse outcome after acute MI. However, our re- sults are consistent with those of previous studies that have reported a significantly higher mortality among the elderly after an acute MI.

Our results also have biologic plausibility because of the pathophysiologic mechanism of the onset of acute coronary syndromes and the physiologic changes associated with aging. Finally, because our study was observational, it is not possible to make inferences regarding the therapies chosen and pa- tient outcomes. Further studies comprising greater numbers of patients will be needed to clarify the ap- propriate therapeutic approach.

We thank Marcia Leavitt for her continued assistance in the management of data in the Myocardial Infarction Registry.

Volume 131, Number 4 American Heart Journal Paul et al. 715

REFERENCES

1. Muller DWM, Topoi EJ. Selection of patients with acute myocardial infarction for thrombolytic therapy. Ann Intern Med 1990;113:949-60.

2. American Heart Association. Heart and Stroke Facts !991. American Heart Association, 1992.

3. Rich MW, Bosner MS, Chung MK, Shen J, McKenzie JP. Is age an in- dependent predictor of early and late mortality in patients with acute myocardial infarction? Am J Med 1992;92:7-13.

4. Hlatky MA, Cotugno HE, Mark DB, O'Connor C, CaliffRM, Prior DB. Trends in physician management of uncomplicated acute myocardial infarction, 1970 to 1987. Am J Cardiol 1988;61:515-18.

5. The BMDP statistical software package, release 7.0 (VAXArMS), Los Angeles, Calif., 1993.

6. ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Randomized trial of intravenous streptekinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial in- farction: ISIS-2. Lancet 1987;1:349-60.

7. Rasheed Q, Hodgson JM. Increased plaque fibrosis and calcification with advancing age in coronary artery disease patients: evidence from intracoronary ultrasound [Abstract]. J Am Coll Cardiol 1993;2 l(2:sup- pl A):481A.

8. Weaver WD, Litwin PE, Martin JS. Effect of age on use ofthrombolytic therapy and mortality in acute myocardial infarction. J Am Coll Car- diol 1991;18:657-62.

9. Working Group on Arteriosclerosis. Arteriosclerosis 1981:report ofthe Working Group on Arteriosclerosis of the National Heart, Lung, and Blood Institute. Publication no. 82-2035. Bethesda, Maryland: Na- tional Institutes of Health, 1981.

10. Smith SC, Gilpin E, Ahnve S. Outlook after acute myocardial infarction in the very elderly compared with that in patients aged 65 to 75 years. J Am Cell Cardiol 1990;16:784-92.

11. Pashos CL, Newhouse J-P, McNeil BJ. Temporal changes in the care and outcomes of elderly patients with acute myocardial infarction, 1987 thr6ugh 1990. JAMA 1993;270:1832-6.

12. Lie JT, Hammond PI. Pathology of the senescent heart: anatomic ob- servations on 237 autopsy studies afpatients 90 to 105 years old. Mayo Clin Proc 1988;63:552-64.

13. Gerstenblith G, Frederiksen J, Yin FCP. Echocardiographic assess- ment of a normal adult aging population. Circulation 1977;56:273-8.

14. Waller BF, Roberts WC. Cardiovascular disease in the very elderly. Analysis of 40 necropsy patients aged 90 years of over. Am J Cardiol 1983;51:403-21.

15. Spirito P, Maron BJ. Influence of aging on Doppler echocardiographic indices of left ventricular diastolic function. Br Heart J 1988;59:672-9.

16. Downes TR, Nomeir AM, Smith KM. Mechanisms of altered pattern of left ventricular filling with aging in subjects without cardiac disease. Am J Cardiol 1989;64:523-7.

17. Avelio AP, Fa-Quan D, Wie-Qiang L. Effects of aging on arterial distensibility in populations with high and low prevalence of hyperten- sion: comparison between urban and rural communities in China. Cir- culation 1985;71:202-10.

18. Wei JY, Gersh BJ. Heart disease in tile elderly. Curr Probl Cardiol 1987;12:7-65.

19. Goldberger AL, RigneyDR. Sudden death is not chaos. In: Krasner S, ed. The ubiquity of chaos. Washington, DC: American Association for the Advancement of Science, 1990:23-34.

20. Geldberger AL, Rigney DR, Mietus J. Nonlinear dynamics in sudden cardiac death syndrome: heart rate oscillations and bifurcations. Experientia 1988;44:983-7.

21. Kleiger RE, Miller JP, Bigger JT Jr, Moss AJ. Multicenter Post-Infarc- tion Research Group. Decreased heart rate variability and its associ- ation with increased mortality after acute myocardial infarction. Am J Cardiol 1987;59:256-62,

22. Pfeffer MA, Weinberg CR, Cook D. Differential changes of autonomic nervous system function with age in man. Am J Med 1983;75:249- 58.