Myocardial Infarction Site and Mortality in Diabetes - Diabetes Care

10 downloads 0 Views 600KB Size Report
No differences in 60-day survival were found between diabetic and nondiabetic subjects ... with regard to the 60-day mortality rate, DIABETES CARE 5.- 31-35 ...
M

yocardial Infarction Site and Mortality in Diabetes

SIMON WEITZMAN, GALEN S. WAGNER, GERARDO HE1SS, THOMAS L. HANEY, AND CECIL SLOME

Survival after a first myocardial infarction (MI) was examined in 54 diabetic and 270 nondiabetic subjects according to anatomic site of MI. Complete survival status information was obtained during a 12-yr follow-up. Compared with nondiabetic subjects, diabetics had a higher proportion of anterior site of MI below the age of 60 in both sexes. A significantly higher mortality was experienced in both sexes by patients with anterior Mis compared with other infarction sites (47% vs. 13% respectively, over 12 yr of follow-up). This excess fatality was differentially distributed by diabetic status. The 60-day mortality following admission with an anterior MI was significantly higher in diabetic (55%) than in nondiabetic subjects (31%). No differences in 60-day survival were found between diabetic and nondiabetic subjects with other infarction sites. Thus, anterior infarct size and diabetic status each convey an increased risk of early postinfarct mortality. Thefindingsfrom this study suggest that the presence of both is synergistic with regard to the 60-day mortality rate, DIABETES CARE 5.- 31-35, JANUARY-FEBRUARY 1982. lthough mortality after myocardial infarction (MI) has been shown to be higher in diabetic than in nondiabetic persons,1"3 causes of this excess risk have not been well established. This high mortality has not been entirely accounted for by comorbid risk factors such as hypertension, obesity, or hyperlipidemia.4*5 Because of the likelihood that the risk factors for the development of coronary artery disease are independent of the risk for dying from an Ml, factors related to the acute coronary event itself may be those that worsen the prognosis for diabetic patients. This study addresses that possibility, attempting to measure the differential mortality in diabetic patients with differing anatomic sites of first acute infarction. MATERIAL AND METHODS

The study population comprised all 487 patients with a first MI admitted to the Duke Medical Center Coronary Care Unit from December 1965 until December 1977. First MI was defined as the absence of previous history and no evidence of previous MI in electrocardiographs on admission or recorded previously. Diagnosis of MI was determined by clinical, enzymatic, and electrocardiographic criteria according to the following definitions. A. Definite Myocardial Infarction 1. Acute: Patients with onset of symptoms within 24 h before admission to the Coronary Care Unit and at least three of the following:

(a) Diagnostic QRS changes on electrocardiograph (b) Total creatinine kinase >130 IU (c) Presence of creatinine kinase myocardial band (MB) by electrophoresis on epi-agar (since 1970 only) (d) Lactic dehydrogenasej greater than lactic dehydrogenase2 B. Infarct Site: Four different groups were considered according to the following criteria: 1. Anterior (a) Initial Q wave of any duration in lead V t and/or (b) Initial R wave in Vj with decreasing R wave in V2 and further in V3 and/or (c) Appearance of Q waves ^0.03-s duration in leads V5 and/or V 6 , AVL, and I 2. Inferior (a) In presence of a negative frontal plane axis, Q wave of any size in both leads AVF and II or (b) Q wave > 0.03 s in lead AVF or (c) Q wave > 0 . 1 m V in AVF, which is >20% of amplitude of subsequent R wave 3. Both: Any combination of components from anterior and inferior 4. Neither: Diagnostic changes in serum enzymes but none of the above QRS changes "Inferior" and "neither" categories were combined into "other" for the purpose of comparing them with "anterior" MI site.)

DIABETES CARE, VOL. 5 NO. 1, JANUARY-FEBRUARY 1982

31

MYOCARDIAL INFARCTION SITE AND MORTALITY IN DIABETES/S. WEITZMAN AND ASSOCIATES

C. Diabetes: Patients were considered diabetic when their medical records documented a diagnosis and treatment of diabetes previous to current admission Of the sample, 121 black patients were excluded because the relatively small number did not permit age, sex, and sitespecific analysis. In addition, 42 white patients categorized as having "both" for sites were excluded because their inclusion would confound the site-specific analysis. This study, therefore, includes the 324 white patients with anterior and "other" infarct sites. Stratification analysis was used to compare diabetic and nondiabetic patients. The Kaplan-Meier Product Limit (PL) Survival Estimate was used to compare the differences in mortality.6 The statistical test used to evaluate these differences was the Mantel Haenszel test for survival data;7 all P values shown are computed from this chi-squared statistic, which incorporates information on time to death. Complete life/death follow-up data are available for the entire study population. RESULTS

The distribution of the study population (324 patients) by age, sex, and presence of diabetes is shown in Table 1. No statistically significant age difference was found between diabetic and nondiabetic patients. Table 2 shows the distribution of MI sites for each age and sex category and diabetic status. In the total sample, anterior and inferior sites are approximately equally respresented (37.3% and 45.1%, respectively). "Neither" sites account for 17.6% of the cases. Figure 1 illustrates the pattern of anterior infarct by age and sex. Among nondiabetic men there are slight but not statistically significant differences in the proportions of anterior sites between the %) than in nondiabetic subjects (31.3%) (P < 0.001) (Figure 3). Two agesex specific groups were found to have the greater risk of death following an anterior myocardial infarction: diabetic men 60 yr or older and diabetic women under 60 yr. As Figure 4 illustrates, 60-day survival differences were not found between diabetic and nondiabetic subjects with other infarct sites. The following three complications of an Ml were considered as potential explanatory factors of the higher mortality among diabetic subjects: (1) sustained hypotension (systolic blood pressure less than 90 mm Hg for at least 1 hr), (2) ventricular arrhythmia, and (3) cardiac arrest (monitored and unmonitored). There were no significant differences in the rates of occurrence of any one of these clinical manifestations between diabetic and nondiabetic subjects following anterior Mis (22.7% and 25.3%, respectively).

he results of the present study confirm the known higher mortality after a documented acute myocardial infarct in diabetic compared with nondiabetic individuals. Although this was observed in the 60day period after admission, the excess mortality was predomi-

30 DAYS

40

50

60

FIG. 3. Sixty-day survival curves following an anterior myocardial infarction in diabetic and nondiabetic subjects. Each circle represents a death. The numbers in parentheses are the numbers of patients alive at each time period. As can be seen, in both diabetic and nondiabetic subjects, the inajority of deaths occurred within the first 20 days.

nantly evident in the first 20 days after the infarction. However, the finding that this excess mortality among diabetic subjects was significant only for those with an anterior MI needs consideration. 100 (151)

90 ^

80

|

70 Non-Diabetics (n = Diabetics (n = 32)

g 60 o cr a. 50 40

DISCUSSION

T

20

0

10

20

30 DAYS

40

50

60

FJG. 4. Sixty-day survival curves following "other" sites of myocardial infarction in diabetic and nondiabetic subjects. Each circle represents a death. The numbers in parentheses are the numbers of patients alive at each time period.

DIABETES CARE, VOL. 5 NO. 1, JANUARY-FEBRUARY 1982

33

MYOCARDIAL INFARCTION SITE AND MORTALITY IN DIABETES/S. WEITZMAN AND ASSOCIATES

To date, no explanations have satisfied the requisite of scientific replication, although a number seem likely to be biologically plausible; issues of selection and information biases may explain the inconsistency of findings. These include the definition and diagnostic criteria used in the ascertainment of diabetes and, in some studies, the nondelineation as to whether the MI was the initial episode or not as well as the selected nature of many samples used in these studies. One major potential bias in this study could be related to the selection of a group of patients with a first Ml who survived long enough to be admitted to a large medical facility. Such preadmission hospital survival may be related to the site of infarction, with excessive preadmission mortality found in cases with anterior myocardial ischemia and infarction.8 The excess in-hospital mortality among those with anterior site infarctions suggests that such selected survival is an unlikely source of bias in this study, although there is no evidence to refute or support this conclusion. The severity of the condition as measured by the proportion of complications following an anterior MI was shown to be similar in diabetic and nondiabetic subjects, thus excluding the plausibility of this explanation as the cause of the observed differentials in mortality. This finding is similar to that reported by others.9'10 This study shows differences in the site of first MI between female diabetic and nondiabetic subjects but not in men. The finding adds to the complexity in the dissection of factors influencing the occurrence of Mis and should be viewed with some caution because of the relatively small number of diabetic patients studied. The difference found is most likely due to the low incidence of anterior MI among nondiabetic young women when compared with all other subgroups. It is possible that young diabetic women develop an "older female" or "male" pattern of site, as the diabetic process introduces cardiovascular changes otherwise associated with aging or with the male sex. However, the high prevalence of anterior MI in diabetic subjects less than 60 yr old cannot be the sole explanation for the differentials in mortality because all diabetic groups show an excess mortality compared with nondiabetic patients, when they had suffered an anterior MI. A study of MI in diabetic patients in Malta has shown a larger proportion of diabetic patients from both sexes with anterior lateral Mis compared with nondiabetic patients. In that study anteroseptal infarctions were also shown to have a worse prognosis in diabetic than in nondiabetic patients.10 The syndrome of an impaired myocardial performance in diabetic patients seems to be a likely explanation of the higher mortality of related anterior MI. First, anterior infarctions tend to be larger11 and hence more disruptive of overall ventricular function. They might, therefore, be expected to exert a greater effect on subsequent prognosis. Second, impaired myocardial function in diabetic patients independently of a degree of cardiac arteriosclerosis has been described. 12~14 This is substantiated by the finding that patients with a familial history of diabetes but no cardiovascular symptoms have a diminished left ventricular end-diastolic and stroke volume and a significant elevation of the left ven-

34

tricular end-diastolic pressure.15 A recent study of asymptomatic young diabetic subjects supported the concept of a specific clinical entity of diabetic cardiomyopathy.16 In summary, two characteristics have been identified in this study population, each carrying an increased risk of early postinfarct mortality: anterior infarct site and diabetic status. It appears that each of these factors contributes independent prognostic significance, and that the presence of both is associated with the highest 60-day mortality rate. This pattern is found in both men and women, with a greater than 3:1 mortality rate ratio for presence of both anterior site and diabetes versus neither. This suggests that there may be a synergistic effect on mortality between the pre-infarct cardiac status induced by diabetes and the well-known greater size of anterior located infarctions, probably complicated by the age and sex of the diabetic patients. These limited but suggestive findings stress the need for further study including measurements of diabetic control and treatment status. ACKNOWLEDGMENTS: The authors acknowledge the help of Steve Samuels, Ph.D., and James W. Fletcher, B.S., in the analyses of the data. From the Diabetes Clinic, Soroka Medical Center and Epidemiology and Health Services Evaluation Units, Ben-Gurion University of the Negev, Beersheva, Israel (S.W.); the Department of Medicine, Cardiology Division, Duke University Medical Center, Durham, North Carolina (G.S.W., T.L.H.), and the Department of Epidemiology, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina (G.H., C.S.). Address reprint requests to Gerardo Heiss, Department of Epidemiology, School of Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27514. REFERENCES 1 Scott, R.: Diabetes and the heart. Am. Heart J. 90: 283-89, 1975. 2 Bradley, R.: In Joslin's Diabetes Mellitus. 11th edit. Marble, A., et al., Ed. Philadelphia, Lea & Febiger, 197.1, chap. 15, p. 417. 3 Stamler, J.: Epidemiology of coronary heart disease. Med. Clin. North Am. 57 (1): 5-46, 1973. 4 Heyden, S., Heiss, G., Bartel, A. G., and Hames, C. G.: Sex differences in coronary mortality among diabetics in Evans County, Ga. Am. J. Chronic Dis. 33: 265-73, 1980. 5 Garcia, M. J., McNamara, P. M., Gorcon, T., and Kannel, W. B.: Morbidity and mortality in diabetics in the Framingham population. Diabetes 23: 105-11, 1974. 6 Kaplan, E. L., and Meier, P.: Nonparametric estimation from incomplete observations. J. Am. Statist. Assoc. 53: 457-81, 1958. 7 Mantel, N.: Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother. Rep. 50: 163-70, 1966. 8 Liberthson, R. R., Nagel, E. L., Hirschman, J. C., Nussenfeld, S. R., Blackbourne, B. D., and Davis, J. H.: Pathophysiologic observations in prehospital ventricular fibrillation and sudden cardiac death. Circulation 49: 790-98, 1974. . 9 Czyzk, A., Krolewski, A. S., Szablowska, S., Alot, A., and Kopczynski, J.: Clinical course of myocardial infarction among diabetic patients. Diabetes Care 3: 526-29, 1980. 10 Zammit Maempel, J. V.: Effect of diabetes on the course of

DIABETES CARE, VOL. 5 NO. 1, JANUARY-FEBRUARY 1982

MYOCARDIAL INFARCTION SITE AND MORTALITY IN DIABETES/S. WEITZMAN AND ASSOCIATES

acute myocardial infarction in Malta. Isr. J. Med. Sci. 14: 424-31, 1978. "Savage, R. M., Wagner, G. S., Ideker, R. E., Podolsky, S. A., and Hackel, D. B.: Correlation of postmortem anatomic findings with electrocardiographic changes in patients with myocardial infarction. Circulation 55: 279-85, 1977. 12 Hamby, R. I., Zoneraich, S., and Sherman, L : Diabetic cardiomiopathy. JAMA 299: 1749-54, 1974. 13 Ledet, T.: Diabetic cardiopathy. Acta Pathol. Microbiol. Scand. Sect. A 84: 421-28, 1976.

14 Regan, T. J., Ahmed, S. S., Haider, B., and Lyons, M. M.: The myocardium and its vasculature in diabetes mellitus (Part II). Mod. Concepts Cardiovasc. Dis. 47:75-78, 1978. 15 Regan, T. J., Lyons, M. M., Ahmed, S. S., Levinson, G. E., Oldewurtel, H. A., Ahmad, M. R., and Haider, B.: Evidence for cardiomyopathy in familial diabetes mellitus. J. Clin. Invest. 60: 885-99, 1977. 16 Sanderson, J. E., Brown, D. J., Rivellese, A., and Kohner, E.: Diabetic cardiomyopathy? An echocardiographic study of young diabetics. Br. Med. J. 1: 404-407, 1978.

DIABETES CARE, VOL. 5 NO. 1, JANUARY-FEBRUARY 1982

35