Circulation
© 1994 American Heart Association, Inc. Volume 90(4), October 1994, pp 1715-1730
Coronary Heart Disease/Myocardial Infarction: Sex Differences in the Management and Long-Term Outcome of Acute Myocardial Infarction: A Statewide Study
[Clinical Investigation and Reports]

Kostis, John B.; Wilson, Alan C.; O'Dowd, Kenneth; Gregory, Patrice; Chelton, Sandra; Cosgrove, Nora M.; Chirala, Anu; Cui, Ting.

For the MIDAS Study Group.
Received April 7, 1994; revision accepted May 31, 1994.
From UMDNJ-Robert Wood Johnson Medical School, New Brunswick, NJ (J.B.K., A.C.W., P.G., S.C., N.M.C., A.C., T.C.) and the New Jersey State Department of Health (K.O.).
Correspondence to John B. Kostis, MD, UMDNJ-Robert Wood Johnson Medical School, One Robert Wood Johnson Pl, CN19, New Brunswick, NJ 08903-0019.

Outline

Graphics

Abstract^

Background We wished to evaluate whether differences in the rate of invasive cardiac procedures between men and women with acute myocardial infarction are associated with different short- and long-term mortality.

Methods and Results The database (Myocardial Infarction Data Acquisition System, MIDAS) included all discharges for the years 1986 and 1987 with the diagnosis of acute myocardial infarction in New Jersey, based on the New Jersey hospital discharge data system (MIDS/UB-82). Accuracy of the data was evaluated by auditing 726 randomly selected charts. The variables examined included age, sex, race, comorbidity (anemia, chronic liver disease, cancer, chronic obstructive pulmonary disease, diabetes, hypertension, prior myocardial infarction), complications (left ventricular dysfunction, arrhythmias, conduction defects), insurance status, performance of cardiac catheterization, percutaneous transluminal coronary angioplasty, or coronary artery bypass graft surgery, and survival up to 3 years. Women were older, had longer hospital stay, and were more likely than men to have anemia, diabetes, hypertension, left ventricular dysfunction, and Medicare or Medicaid insurance coverage. They were less likely than men to be admitted to a hospital equipped to perform invasive procedures or to have chronic obstructive pulmonary disease, chronic liver disease, prior myocardial infarction, or arrhythmias. After adjustment for these differences, women were less likely than men to have cardiac catheterization. Cardiac catheterization was associated with lower mortality. Women up to age 70 had higher 3-year death rates than men after adjustment for age, race, comorbidity, complications, and insurance type. This difference between men and women was somewhat diminished after the performance of cardiac catheterization and revascularization was taken into account. Unadjusted mortality was high in this study group.

Conclusions Women with acute myocardial infarction are less likely to have invasive cardiac procedures and have higher 3-year adjusted death rate up to age 70 than men. (Circulation. 1994;90:1715-1730.)



Women are less likely than men to have diagnostic and therapeutic procedures for coronary artery disease [1,2]. Use of cardiac catheterization, percutaneous transluminal coronary angioplasty (PTCA), and aortocoronary bypass graft surgery (CABG) in women after acute myocardial infarction (MI) has been reported to be lower than [3,4,5,6,7,8] or equal to [9] that in men. These reports did not establish whether differences in use of procedures reflect appropriate clinical practice considering other differences in clinical characteristics between the sexes and whether subsequent mortality in women is adversely affected by the lower rate of use.

The purpose of this study was to assess, in a statewide MI database audited for accuracy, whether women with acute MI have lower rates of diagnostic and therapeutic invasive cardiac procedures and whether different rates of use of procedures are associated with different 3-year survival in men and women.

Methods^

The study was based on the Myocardial Infarction Data Acquisition System (MIDAS), a statewide database that includes information on all patients (n=49 250) who were discharged from 90 nonfederal hospitals in New Jersey in the years 1986 and 1987 with the diagnosis of acute MI. Abstracted discharge data were obtained from the New Jersey hospital discharge data system (MIDS/UB-82). This system, developed to support the hospital rate-setting process, provides a unique opportunity to study cardiovascular diseases throughout the state. The confidentiality of hospital patient records was maintained, and the entire study was approved by the Institutional Review Board.

Inclusion Criteria and Variables^

Hospital discharge records from the 1986 and 1987 MIDS/UB-82 files were included in the MIDAS database if an International Classification of Diseases 9th Revision (ICD-9) code in the range 410.0-410.9 (ICD-9 410) was present in any of the nine discharge diagnosis fields, ie, when the diagnosis was the principal diagnosis or another diagnosis. The discharge records included current procedural terminology codes for procedures performed and other relevant patient information such as age, sex, race, insurance status, and admission and discharge dates.

Also included in this study were composite variables indicating the presence or absence of diabetes (ICD-9 250.00-250.90), hypertension (ICD-9 401.00-405.99), chronic obstructive pulmonary disease (ICD-9 490.00-496.99), chronic renal disease (ICD-9 580.00-589.99), chronic liver disease (ICD-9 571.00-571.99), anemia (ICD-9 280.00-281.30, 281.90, 282.00, 283.00-285.99), cancer (ICD-9 140.00-208.00), and cerebrovascular disease (ICD-9 430.00-438.99). Two composite indices of mechanical or electrical complications of MI were generated. The first, left ventricular dysfunction (LVD), included presence of any of the following: old MI (ICD-9 412.00), congestive heart failure (CHF) (ICD-9 428.00), left heart failure (ICD-9 428.10), cardiomegaly (ICD-9 429.30), alcoholic cardiomyopathy (ICD-9 425.50), rupture of papillary muscle (ICD-9 429.60), rupture of chordae tendineae (ICD-9 429.50), acquired cardiac septal defects (ICD-9 429.71), hypertensive heart disease with CHF (ICD-9 402.91), cardiogenic shock (ICD-9 785.51), or ventricular aneurysm (ICD-9 414.10). The second, electrical instability (arrhythmia), included presence of cardiac dysrhythmias (ICD-9 427.00), complete atrioventricular block (ICD-9 426.00), unspecified atrioventricular block (ICD-9 426.10), Mobitz (type II) atrioventricular block (ICD-9 426.12), left bundle branch block (ICD-9 426.20), other left bundle branch block (ICD-9 426.30), right bundle branch block (ICD-9 426.40), other and unspecified bundle branch block (ICD-9 426.50), other heart block (ICD-9 426.6), or other specified and unspecified conduction disorders (ICD-9 426.90).

Exclusions^

Because of missing century data for date of birth, we excluded records with apparent ages 99 years (<0.1%). For every patient, only the first recorded (index) infarction that occurred in this database was considered. Multiple discharges for a given person were identified by internal record linkage. Excluding second and subsequent records for the same patient within the 2 years resulted in the elimination of 6655 (13.5%) of the discharges, leaving 42 595 upon which the present analysis is based.

Patients with acute MI who were hospitalized in another state are not included in this analysis. In addition, the 11 488 persons who died in New Jersey in 1986 and 1987 from possible acute MI (identified on the death certificate by the single cause of death code 410) before hospital admission are not included in this study. In addition, patients who had an MI and were admitted to federal hospitals or nursing homes are not included in the MIDAS database. These patients probably account for <3% of the cases.

Data Auditing^

A random sample of charts was audited to verify the presence or absence of acute MI as well as the accuracy of other information included in the database. Auditing of the data was performed by comparing the information from the UB-82 files with that written in the hospital charts. Seven hundred twenty-six charts randomly chosen from patients admitted in 12 hospitals were abstracted and reviewed. The hospitals and charts were chosen by an algorithm designed to give each of the 42 595 charts an equal chance of being audited. Of these 12 hospitals, 3 had facilities for cardiac catheterization, and 1 offered revascularization procedures. Of the 726 charts selected for auditing, 669 represented first infarctions in the years 1986 and 1987.

Criteria for Diagnosis of Acute MI^

The presence or absence of acute MI was ascertained from criteria modified from the beta -Blocker Heart Attack Trial, based on three pieces of information: symptoms, cardiac enzyme levels, and ECG evidence [10]. On the basis of this information, each audited chart was classified as indicating definite, probable, recent, or absent MI.

A chart was classified as "definite acute MI" when at least two of the following three criteria were present: (1) definite Q waves (Minnesota Code 1-1-1 through 1-2-7 for Q-QS patterns) developed during the hospitalization, (2) elevated cardiac enzymes as defined below, or (3) symptoms of acute MI associated with newly developed ST elevation (Minnesota Code 9-2) or major T-wave inversion (Minnesota Code 5-1 to 5-2) or ST depression (Minnesota Code 4-1 to 4-2).

The MI was classified as "probable" when (1) typical symptoms, elevated enzymes, and minor ST-(Minnesota Code 4-3 to 4-4) or T-wave changes (5-3 to 5-4) were present in the absence of definite Q waves, or major ST- or T-wave findings (described in 1 and 3 above) on the ECG; (2) serum enzymes were not recorded within 3 days of admission, but there was evidence of typical symptoms as well as the major ST or T abnormalities mentioned in (3) above; or (3) when typical symptoms were not recorded but there was evidence of both enzyme elevation and ECG abnormalities indicating either minor Q-wave (Minnesota Code 1-3-1 to 1-3-6) or major ST- or T-wave changes mentioned in (3) above.

"Recent" MI was diagnosed when the criteria mentioned under definite or probable MI were present but it could be ascertained from the chart that the infarction had occurred before the index admission but within the past 3 months and verified in the present admission by the presence of continuing Q waves.

When a record could not be classified as indicating definite, probable, or recent infarction, it was classified as having no evidence of infarction ("absent").

The presence of chronic left bundle branch block (ICD-9 7-1-1), complete heart block (6-1), Wolff-Parkinson-White syndrome (6-4-1 or 6-4-2), pacemaker (6-8), ventricular fibrillation (8-2-1), or persistent ventricular rhythm (8-2-2 or 8-2-3) invalidated ECG readings. ECGs were coded according to the Minnesota Code and coded on forms similar to those developed by the Atherosclerosis Risk in Communities (ARIC) Study [11].

Enzyme elevation was considered to be present when measurements within the first 3 days of admission showed an elevation of total creatine kinase, serum glutamic-oxaloacetic transaminase, or lactate dehydrogenase beyond two times the upper normal for the laboratory and the creatine kinase MB fraction was >5%, or "present."

Other Information and Vital Status^

Age was considered to be correct if it was within 364 days of the true age of the patient as verified by the chart. All other variables were considered correct if the entries in the database corresponded exactly to the information in the chart.

Vital status follow-up of all patients was performed by matching the MIDAS database with the 1986-1990 single cause of death data files at the New Jersey State Department of Health by use of a specially designed automated record linkage system provided by DataStar, Inc. The methodology is described elsewhere [12]. Each patient was followed for a 3-year period from admission or until death if death occurred before 3 years. In-hospital mortality was validated from the 726 audited charts. The accuracy of the record linkage system matching MIDAS and death certificate records was tested in the total cohort (42 595) by in-hospital mortality. The performance of cardiac catheterization, PTCA, and CABG after the index hospitalization was ascertained by matching the MIDAS database with the 1986-1990 UB-82 records for these procedures in New Jersey.

Statistical Analysis^

Clinical and demographic characteristics of the groups of interest were compared by Pearson's chi squared (chi2) test for categorical variables and Student's t test for continuous variables [13]. The cumulative incidence of patients who received cardiac catheterization, angioplasty, and CABG were described for men and women, compared by relative risks calculated from the ratio of the two cumulative incidence measures, and stratified by potential confounding variables [14]. Odds ratios adjusted for the potential confounding factors were presented with 95% confidence intervals based on log-linear regression using the maximum-likelihood method [15]. Survival curves were generated based on the Kaplan-Meier method [16]. Plots of the survival curves and the log-log transformation of the survival curves were used to study the survival in subgroups of interest and to evaluate the proportionality assumption for the proportional-hazards regression model. If the proportionality assumption did not hold true for a given covariate, then certain subgroups were excluded or analyzed separately. Proportional- hazards regression was used to assess the relative mortality rates of men compared with women over the 3-year follow-up period while potential confounding variables were adjusted for [17]. Statistical analyses were performed with base sas and sas/stat procedures of the SAS System (SAS Institute, Inc) version 6.07 run on a Hewlett Packard 9000/845 computer. Analyses were done with data from all patient cases and then repeated including only those cases in which acute MI was the principal diagnosis.

Results^
Population Under Study^

During the years 1986 and 1987, 49 250 patients with acute MI as a diagnosis were discharged from 90 nonfederal hospitals in the state of New Jersey. When only the first hospitalization for each patient was considered, there were 42 595 discharges. Hospitalizations with acute MI as the primary diagnosis accounted for 32 413 cases (76.1%). Women (17 422) composed 40.9% of the study group and 38.9% of those with acute MI as the principal diagnosis.

Validation of the Diagnosis of Acute MI and Related Variables^

The audited sample was representative of the total cohort with respect to age, comorbid conditions, prior history of MI, complications, length of stay, and insurance status. The percentage of patients of white race (non- Hispanic) was higher in the audited sample, whereas the percentage of patients in health maintenance organizations and the frequencies of cardiac catheterization and CABG were lower in the audited set Table 1.



Table 1. Comparison of the Audited Sample to the Total Cohort of Patients Discharged in 1986 and 1987 in New Jersey With a Diagnosis of Acute Myocardial Infarction (ICD-9 410)

A definite acute MI was verified in 449 (67.1%) of the 669 charts audited, a probable acute MI in 111 (16.6%), and a recent MI in 46 (6.9%). The diagnosis of "definite," "probable," or "recent" MI could not be supported by the chart in 63 cases (9.4%). This percentage was lower (6.4%) when only the charts with acute MI as the principal diagnosis were considered Table 2. The percentage of patients in whom a diagnosis of acute MI could not be supported by the chart was higher in women than in men, both in the total audited set (11.7% versus 8.1%) and in the subset of audited charts with acute MI as the primary diagnosis (9.0% versus 4.8%).



Table 2. Validation of the Diagnosis of Acute Myocardial Infarction*

The UB-82 file information was correct in 99.3% (664/669) of the audited cases for classification of sex, 99.7% (667/669) for age, 98.8% (661/669) for race, 99.3% (664/669) for length of stay, 98.8% (661/669) for vital status at discharge, 99.9% (668/669) for the performance of cardiac catheterization, and 99.7% (667/669) for the admission date. The sensitivity and specificity of the MIDAS database in detecting in-hospital mortality were 96.8% and 100.0%, respectively. The sensitivity and specificity of the algorithm matching MIDAS to the death tapes were 98.16% and 98.17%, respectively.

Clinical Characteristics^

The average age of the patients was 67.8+-13.1 years Table 3. Approximately 87% were classified as white. In all, 8.1% had a history of prior MI, 33.1% had hypertension, 22.8% diabetes mellitus, 8.7% anemia, and 11.6% chronic pulmonary disease. Approximately 40% of the patients had entries on their discharge summary indicating electrical or mechanical complications (arrhythmia or LV dysfunction). Women were more likely to have subendocardial infarction as a discharge diagnosis (21.7% versus 16.9%). Women were older on average by approximately 7.5 years and stayed in hospital longer. They were more likely to have anemia, diabetes, hypertension, or entries indicating left ventricular dysfunction, Medicare, or Medicaid. Women were less likely than men to have chronic liver or pulmonary disease, prior MI, or entries indicating electrical instability and commercial or HMO insurance coverage. Women were also less likely to be admitted to a hospital offering revascularization procedures or equipped with a cardiac catheterization laboratory.



Table 3. Selected Clinical Characteristics of Men and Women Discharged in 1986 and 1987 From New Jersey Hospitals With a Diagnosis of Acute Myocardial Infarction (ICD-9 410)

Invasive Procedures in Men and Women^

Women were less likely than men to undergo cardiac catheterization (7.2% versus 12.9%), PTCA (1.1% versus 2.1%), or CABG (1.8% versus 2.5%) during their admission Table 3. For procedures performed within 3 months of admission, the respective rates were 15.5% (catheterization), 2.8% (PTCA), and 4.7% (CABG) for women and 28.6%, 5.7%, and 8.3% in men, respectively. When the entire study period was considered, the rates were 18.4%, 3.5%, and 6.0% in women and 32.3%, 6.9%, and 10.4% in men, respectively. The lower rate of use of procedures in women was observed consistently in different age strata, in the presence and absence of comorbid conditions and complications, and in different insurance coverage types and races (Table 4A through 4C).



Table 4A. Procedure Rates During Index Hospitalization in Men and Women After Stratification for Potential Confounding Variables ----- Table 4B. Procedure Rates Within 3 Months of Index Hospitalization in Men and Women After Stratification for Potential Confounding Variables ----- Table 4C. Procedure Rates During Entire Study Period in Men and Women After Stratification for Potential Confounding Variables

Among patients who had cardiac catheterization in the index admission, the percentage having PTCA at that admission was similar in men and in women (13.4% and 12.1%, respectively). The rate of CABG in patients who had cardiac catheterization was 16.0% in men and 11.9% in women. When the first 3 months were considered, the rates of PTCA were 18.3% in women and 21.0% in men and the rates for CABG, 30.3% and 30.4%, respectively. The corresponding rates during the entire study period were 19.2% and 21.6% for PTCA and 32.4% and 32.3% for CABG. When cardiac catheterization was performed during the first week of hospitalization, the rates for PTCA were 19.3% in men and 13.5% in women, and the rates for CABG were 20.8% and 18.2%, respectively.

At the time of the study (1986 through 1987), 24 of the 90 hospitals provided coronary arteriography, and nine of these performed revascularization procedures.

Multiple logistic regression using sex, age, comorbidities, complications, race, and insurance coverage identified sex as an independent predictor of cardiac catheterization and indicated that men had a 37% higher likelihood of cardiac catheterization Table 5. Left ventricular dysfunction, electrical instability, diabetes, liver disease, chronic obstructive pulmonary disease, and increasing age were also associated with lower likelihood of cardiac catheterization Table 5.



Table 5A. Adjusted Odds Ratios for Use of Cardiac Catheterization During the Index Hospitalization in Patients Discharged in 1986 and 1987 in New Jersey With a Diagnosis of Acute Myocardial Infarction (ICD-9 410)* ----- Table 5B. Adjusted Odds Ratios for Use of Cardiac Catheterization Within 3 Months From Index Hospitalization in Patients Discharged in 1986 and 1987 in New Jersey With a Diagnosis of Acute Myocardial Infarction (ICD-9 410)* ----- Table 5C. Adjusted Odds Ratios for Use of Cardiac Catheterization During the Entire Study Period in Patients Discharged in 1986 and 1987 in New Jersey With a Diagnosis of Acute Myocardial Infarction (ICD-9 410)*

Mortality After Acute MI^

Short- and long-term mortality of hospitalized patients with acute MI was high. Mortality was 5.4% on day 1, 12.6% on day 7, 26.7% at 1 month, 30.4% at 6 months, and 33.4%, 39.9%, and 43.7% at 1, 2, and 3 years, respectively. The great majority (91.4%) of deaths in this time frame were attributed to cardiovascular causes. In-hospital mortality was higher for women than men (23.8% versus 17.0%). Older patients and patients with the composite left ventricular dysfunction (LVD) index had higher mortality Figure 1. Cardiac catheterization was associated consistently with lower mortality in both men and women Figure 2.



Figure 1. Bar graph showing effect of left ventricular dysfunction on 3-year mortality in men and women stratified by age group. The composite index, LVD (shaded bars), was present in 7995 women; 6200 women had no complication (open bars). In men, 9841 had LVD, and 9771 had no complication



Figure 2. Survival plots showing the effect of cardiac catheterization in men (catheterized, top two lines; not catheterized, bottom two lines) and women hospitalized for acute myocardial infarction in 1986 and 1987 in New Jersey. A, Catheterization done during index admission. B, Catheterization done within 3 months of admission. C, Catheterization done at any time during the study period

Mortality rates for all patients were evaluated by multivariate proportional-hazards regression. To maintain the proportionality assumption of this analysis, three age groups (30 to 49, 50 to 69, and 70 to 89 years) were studied, and patients with cancer (3.2%) and nephritis (5.0%) were excluded. Age, left ventricular dysfunction, and diabetes were identified as factors independently associated with a higher mortality rate (Table 6A through 6C). In both men and women, cardiac catheterization was associated with lower mortality. Survival analysis revealed that women had lower unadjusted survival rates than men Table 7. This was statistically significant up to age 70 years. Women had a 69% higher death rate in the age group 30 to 49 years, a 21% higher death rate in the age group 50 to 69 years, and a 1% higher rate in the oldest age group. After adjustment for 11 covariates including age, race, comorbidity, complications, and insurance type, singly and in combination, death rates were still higher (although to a lesser extent) in women up to age 70 years, whereas after this age men had higher death rates. The magnitude of difference between the adjusted and unadjusted death rate ratios was greatest in the youngest age group and least in the oldest. If adjustment is made for the performance of cardiac catheterization (or cardiac catheterization and PTCA and CABG) in addition to the other 11 variables, the death rate ratio between men and women up to age 70 years becomes less pronounced and is no longer statistically significant. Similar findings were observed when only cardiovascular mortality was studied and when either total or cardiovascular mortality was examined in the group of patients who had acute MI as the primary diagnosis.



Table 6A. Analysis of Mortality Based on 3-Year Follow-Up for Patients Discharged in 1986 and 1987 in New Jersey With a Diagnosis of Acute Myocardial Infarction (ICD-9 410)* (Cardiac Catheterization Performed During Index Hospitalization) ----- Table 6B. Analysis of Mortality Based on 3-Year Follow-Up for Patients Discharged in 1986 and 1987 in New Jersey With a Diagnosis of Acute Myocardial Infarction (ICD-9 410)* (Cardiac Catheterization Performed Within 3 Months of Index Hospitalization) ----- Table 6C. Analysis of Mortality Based on 3-Year Follow-Up for Patients Discharged in 1986 and 1987 in New Jersey With a Diagnosis of Acute Myocardial Infarction (ICD-9 410)* (Cardiac Catheterization Performed During the Entire Study Period)



Table 7. Effect of Adding Covariates on Death Rate Ratios for Sex Calculated by Cox Proportional- Hazards Analysis*

Discussion^

Older age, more severe or complicated disease, and differences in comorbid states, socioeconomic status, insurance coverage, presenting symptomatology, and the predictive value of stress tests as well as sex bias have all been proposed as causes of the lower use of procedures in women that has been observed in most studies [1,2,3,4,5,6,7,8,18]. Higher complication rates, higher hospital mortality, and lower success rates of interventions may be additional factors leading to avoidance of procedures in some women [19,20,21,22,23,24,25,26,27,28,29].

The present study provides additional data indicating that the rate of use of invasive cardiovascular procedures after an acute MI is lower in women and that this difference cannot be completely explained by factors such as age, comorbidity, complications, or insurance type.

Our study extends previous findings by reporting on almost all patients who had MI in a given geographic area and by having established the accuracy of the data and the diagnosis of acute MI by chart review of a random sample of the cases. In our study, the diagnosis of acute MI could not be supported by chart audit in a higher percentage of women than in men. This may offer an additional reason for differences in use, if physicians treat patients with "diagnosed but not proven" MI less aggressively.

The data presented here support the majority of previous reports indicating lower use of invasive procedures in women with acute MI. One study, from a single institution and potentially of inadequate statistical power, reported no difference between men and women in the rate of coronary arteriography after MI [9].

Recognizing that women are less likely to receive cardiac catheterization and revascularization after an acute MI, it is not known whether this represents underuse in women or overuse in men. Our data on survival may shed light on this issue. Cardiac catheterization was associated with lower mortality in both men and women. In the absence of randomization, it is difficult to ascertain whether the better outcome with this invasive strategy is due either to selection of low-risk patients for the procedure or to a beneficial effect of an intervention such as PTCA or CABG that was done on the same or subsequent admission on patients who had catheterization. However, the persistence of the lower risk after adjustment for factors known to influence prognosis, including age, comorbidity, and complications, suggests that the interventions may have a beneficial effect. Randomized clinical trials have shown that an invasive strategy after infarction is appropriate only when required for clinical indications [30,31]. In SAVE, coronary arteriography and revascularization were performed more commonly in US than in Canadian participating hospitals, but mortality was not different. The rate of cardiac catheterization in SAVE was much higher than in the present study [32]. In this study, the higher adjusted death rate of women compared with men up to age 70 years may suggest that the underuse of (beneficial) procedures is associated with a less favorable outcome in women. The fact that the difference in death rates between men and women is diminished after adjustment for the performance of invasive procedures implies that the association between sex and death rate was confounded by the different rates of performance of cardiac catheterization. A higher long-term mortality in women after acute MI has been reported by previous investigators [33,34,35]. Other studies have indicated a poorer long-term survival in men [36,37,38,39]. In the present study, a higher adjusted death rate was observed in men only in the oldest age group (70 to 89 years). This is consonant with a recent report in the Medicare population in which adjusted mortality was higher in men only in the age groups >74 years [5]. In this age group, the rates for cardiac catheterization were low, and factors other than the 11 variables used for adjustment may become important. The low rate of cardiac catheterization and PTCA (and CABG) in both men and women may potentially be insufficient to markedly affect overall prognosis in the total cohort. In any case, overuse of invasive procedures in men with acute MI based on 3-year survival rates is not supported by the data presented. On the contrary, a significant underuse in women (compared with men) is evident, and a suggestion is present that it is associated with adverse long-term outcomes.

Alternatively, in the majority of patients, a significant long-term benefit might not derive from these procedures, which may instead be markers of better prognosis unrelated to the confounding variables considered in our analysis. Cardiac catheterization may be associated with improved prognosis by allowing revascularization procedures in the index or subsequent hospital admission, by facilitating better medical management, or by being related to better physician and patient characteristics.

The conclusions of this study are strengthened by the fact that similar results are obtained when diagnostic and therapeutic procedures performed during the admission for the index infarction or at 3 months or throughout the follow-up were considered. The retrospective nature of the study and the lack of quantitative measures of coronary anatomy, LV function, risk factors, thrombolytic therapy, and postdischarge management are limitations of the study. In addition, the study pertains to practice more than 5 years ago.

Nevertheless, our data indicate that women with acute MI are less likely to have interventions than men. This may lead to less favorable long-term outcome, especially in the younger age groups (up to 70 years). Invasive procedures are associated with better prognosis, although this may be due in part to selection of low-risk patients for these procedures.

Acknowledgments^

This study was supported in part by grants from the Robert Wood Johnson Foundation (14148) and the New Jersey State Department of Health and the UMDNJ-Robert Wood Johnson Medical School Cardiovascular Institute and Detwiler Fund. We wish to thank New Jersey Health Commissioners Molly Coye, MD, and Bruce Siegel, MD, for their support; Drs Joseph Fleiss and Shaw-Hwa Lo of Columbia University School of Public Health for their help in structuring the sampling strategy of the audited subset; and Lawrence A. Meinert, MD, and the members of the New Jersey Commissioner of Health's Cardiac Services Subcommittee on Quality Audit Systems and Volume Standards (Harvey V. Fineberg, MD, Peter Frommer, MD, John Gregory, MD, John B. Kostis, MD, Victor Parsonnet, MD, Nolan B. Sommer, PhD, Alan B. Cohen, ScD, and John Gontarski, MCRP).

REFERENCES^

1. Tobin JN, Wassertheil-Smoller S, Wexler JP, Steingart RM, Budner N, Lense L, Wachspress J. Sex bias in considering coronary bypass surgery. Ann Intern Med. 1987;107:19-25. Bibliographic Links [Context Link]

2. Khan SS, Nessim S, Gray R, Czer LS, Chaux A, Matloff J. Increased mortality of women in coronary artery bypass surgery: evidence for referral bias. Ann Intern Med. 1990;112:561-567. Bibliographic Links [Context Link]

3. Ayanian JZ, Epstein AM. Differences in the use of procedures between women and men hospitalized for coronary heart disease. N Engl J Med. 1991;325:221-225. Bibliographic Links [Context Link]

4. Wenger NK. Gender, coronary artery disease, and coronary bypass surgery. Ann Intern Med. 1990;112:557-558. [Context Link]

5. Udvarhelyi IS, Gatsonis C, Epstein AM, Pashos CL, Newhouse JP, McNeil BJ. Acute myocardial infarction in the medicare population: process of care and clinical outcomes. JAMA. 1992; 268: 2530-2536. Bibliographic Links [Context Link]

6. Maynard C, Litwin PE, Martin JS, Weaver WD. Gender differences in the treatment and outcome of acute myocardial infarction: results from the Myocardial Infarction Triage and Intervention Registry. Arch Intern Med. 1992;152:972-976. Bibliographic Links [Context Link]

7. Chiriboga DE, Yarzebski J, Goldberg RJ, Chen Z, Gurwitz J, Gore JM, Alpert JS, Dalen JE. A community-wide perspective of gender differences and temporal trends in the use of diagnostic and revascularization procedures for acute myocardial infarction. Am J Cardiol. 1993;71:268-273. Bibliographic Links [Context Link]

8. Steingart RM, Packer M, Hamm P, Coglianese ME, Gersh B, Geltman EM, Sollano J, Katz S, Moye L, Basta LL, Lewis SJ, Gottlieb SS, Bernstein V, McEwan P, Jacobson K, Brown EJ, Kukin ML, Kantrowitz NE, Pfeffer MA, for the Survival and Ventricular Enlargement Investigators. Sex differences in the management of coronary artery disease. N Engl J Med. 1991;325: 226-230. [Context Link]

9. Krumholz HM, Douglas PS, Lauer MS, Pasternak RC. Selection of patients for coronary angiography and coronary revascularization early after myocardial infarction: is there evidence for a gender bias? Ann Intern Med. 1992;116:785-790. Bibliographic Links [Context Link]

10. Beta-Blocker Heart Attack Trial Research Group. A randomized trial of propranolol in patients with acute myocardial infarction. JAMA. 1982;247:1707-1714. [Context Link]

11. The ARIC Investigators. The Atherosclerosis Risk in Communities (ARIC) study: design and objectives. Am J Epidemiol. 1989;129: 687-702. Bibliographic Links [Context Link]

12. Jaro MA. Advances in record-linkage methodology as applied to matching the 1985 census of Tampa, Florida. J Am Stat Assoc. 1989;84:414-420. [Context Link]

13. Snedecor G, Cochran W. Statistical Methods. 7th ed. Ames, Iowa: Iowa State University Press; 1980:507. [Context Link]

14. Kleinbaum D, Kupper L, Morgenstern H. Epidemiologic Research: Principles and Quantitative Methods. New York, NY: Van Nostrand Reinhold; 1982:529. [Context Link]

15. Hosmer D, Lemeshow S. Applied Logistic Regression. New York, NY: John Wiley & Sons, Inc; 1989:307. [Context Link]

16. Kahn H, Sempos C. Statistical Methods in Epidemiology. New York, NY: Oxford University Press; 1989:292. [Context Link]

17. Selvin S. Statistical Analysis of Epidemiologic Data. New York, NY: Oxford University Press; 1991:375. [Context Link]

18. Sketch MH, Mohiuddin SM, Lynch JD, Zenchka AE, Runco V. Significant sex differences in the correlation of electrocardiographic exercise testing and coronary arteriograms. Am J Epidemiol. 1989;129:687-702. [Context Link]

19. Douglas JS, King SB III, Jones EL, Craver JM, Bradford JM, Hatcher CR Jr. Reduced efficacy of coronary bypass surgery in women. Circulation. 1981;64(suppl II):II-11-II-16. Ovid Full Text [Context Link]

20. Johnson LW, Lozner EC, Johnson S, Krone R, Pichard AD, Ventrovec GW, Noto TJ and the Registry Committee of the Society for Cardiac Angiography and Interventions. Coronary arteriography 1984-1987: a report of the registry of the society for cardiac angiography and intervention, I: results and complications. Cathet Cardiovasc Diagn. 1989;17:5-10. [Context Link]

21. Dorros G, Cowley MJ, Janke L, Kelsey SF, Mullin SM, Raden MV. In-hospital mortality rate in the National Heart, Lung, and Blood Institute Percutaneous Transluminal Coronary Angioplasty Registry. Am J Cardiol. 1984;53:17C-21C. Bibliographic Links [Context Link]

22. Cowley MJ, Mullin SM, Kelsey SF, Kent KM, Gruentzig AR, Detre KM, Passamani ER. Sex differences in early and long-term results of coronary angioplasty in the NHLBI PTCA registry. Circulation. 1985;71:90-97. Ovid Full Text Bibliographic Links [Context Link]

23. Kent KM, Bentivoglio LG, Block PC, Bourassa MG, Cowley MJ, Dorros G, Detre KM, Gosselin AJ, Gruentzig AR, Kelsey SF, Mock MB, Mullin SM, Passamani ER, Myler RK, Simpson J, Stertzer SH, Van Raden MJ, Williams DO. Long-term efficacy of percutaneous transluminal coronary angioplasty (PTCA): report from the National Heart, Lung, and Blood Institute PTCA Registry. Am J Cardiol. 1984;53:27C-31C. Bibliographic Links [Context Link]

24. Bell MB, Holmes DR Jr, Berger PB, Garratt KN, Bailey KR, Gersh BJ. The changing in-hospital mortality of women undergoing percutaneous transluminal coronary angioplasty. JAMA. 1993;269:2091-2095. Ovid Full Text Bibliographic Links [Context Link]

25. Casale PN, Whitlow PL, Franco I, Grigera F, Pashkow FJ, Topol EJ. Comparison of major complication rates with new atherectomy devices for percutaneous coronary intervention in women versus men. Am J Cardiol. 1993;71:1221-1223. Bibliographic Links [Context Link]

26. Fisher LD, Kennedy JW, Davis KB, Maynard C, Fritz JK, Kaiser G, Myers WO, and the Participating CASS Clinics. Association of sex, physical size, and operative mortality after coronary artery bypass in the Coronary Artery Surgery Study (CASS). J Thorac Cardiovasc Surg. 1982;84:334-341. Bibliographic Links [Context Link]

27. Wenger NK, Speroff L, Packard B. Cardiovascular health and disease in women. N Engl J Med. 1993;329:247-256. Bibliographic Links [Context Link]

28. Loop FD, Golding LR, Macmillan JP, Cosgrove DM, Lytle BW, Sheldon WC. Coronary artery surgery in women compared with men: analyses of risks and long-term results. J Am Coll Cardiol. 1983;1:383-390. Bibliographic Links [Context Link]

29. Kelsey SF, James M, Holubkov AL, Holubkov R, Cowley MJ, Detre KM, and Investigators from the National Heart, Lung, and Blood Institute Percutaneous Transluminal Coronary Angioplasty Registry. Results of percutaneous transluminal coronary angioplasty in women: 1985-1986 National Heart, Lung, and Blood Institute's Coronary Angioplasty Registry. Circulation. 1993;87: 720-727. [Context Link]

30. The TIMI Study Group. Comparison of invasive and conservative strategies after treatment with intravenous tissue plasminogen activator in acute myocardial infarction: results of the thrombolysis in myocardial infarction (TIMI) phase II trial. N Engl J Med. 1989;320:618-627. Bibliographic Links [Context Link]

31. SWIFT (Should We Intervene Following Thrombolysis?) Trial Study Group. SWIFT trial of delayed elective intervention v conservative treatment after thrombolysis with anistreplase in acute myocardial infarction. BMJ. 1991;302:555-560. Bibliographic Links [Context Link]

32. Rouleau JL, Moye LA, Pfeffer MA, Arnold JMO, Bernstein V, Cuddy TE, Dagenais GR, Geltman EM, Goldman S, Gordon D, Hamm P, Klein M, Lamas GA, McCans J, McEwan P, Menapace FJ, Parker JO, Sestier F, Sussex B, Braunwald E for the SAVE Investigators. A comparison of management patterns after acute myocardial infarction in Canada and the United States. N Engl J Med. 1993;328:779-784. Bibliographic Links [Context Link]

33. Greenland P, Reicher-Reiss H, Goldbourt U, Behar S, Israeli SPRINT Investigators. In-hospital and 1-year mortality in 1,524 women after myocardial infarction: comparison with 4,315 men. Circulation. 1991;83:484-491. [Context Link]

34. Tofler GH, Stone PH, Muller JE, MILIS Study Group. Effects of gender and race on prognosis after myocardial infarction: adverse prognosis for women, particularly black women. J Am Coll Cardiol. 1987;9:473-482. Bibliographic Links [Context Link]

35. Kannel WB, Sorlie P, McNamara PM. Prognosis after initial myocardial infarction: the Framingham study. Am J Cardiol. 1979;44: 53-59. Bibliographic Links [Context Link]

36. Goldberg RJ, Gorak EJ, Yarzebski J, Hosmer DW, Dalen P, Gore JM, Alpert JS, Dalen JE. A communitywide perspective of sex differences and temporal trends in the incidence and survival rates after acute myocardial infarction and out-of-hospital deaths caused by coronary heart disease. Circulation. 1993;87:1947-1953. [Context Link]

37. Merrilees MA, Scott PJ, Norris RM. Prognosis after myocardial infarction: results of 15 year follow-up. BMJ. 1984;288:356-359. [Context Link]

38. Pohjola S, Siltanen P, Romo M. Five-year survival of 728 patients after myocardial infarction: a community study. Br Heart J. 1980; 43:176-183. Bibliographic Links [Context Link]

39. Martin CA, Thompson PL, Armstrong BK, Hobbs MS, DeKlerk N. Long-term prognosis after recovery from myocardial infarction: a nine-year follow-up of the Perth Coronary Register. Circulation. 1983;68:961-969. [Context Link]



Accession Number: 00003017-199410000-00015