Skip Navigation U.S. Department of Health and Human Services
Agency for Healthcare Research Quality
Archive print banner

Feature Story

This information is for reference purposes only. It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work. Persons with disabilities having difficulty accessing this information should contact us at: Let us know the nature of the problem, the Web address of what you want, and your contact information.

Please go to for current information.

New EKG add-on could reduce deaths from heart attack

Busy emergency room doctors now have a new ally in their fight to save the lives of heart attack victims. The ally is a computerized decisionmaking aid that will be built into the standard electrocardiograph (EKG) machine to instantly print predictions, based on the patients EKG and other clinical information, of whether that individual is likely to benefit from potentially lifesaving treatment with thrombolytic (clot-dissolving) drugs.

The new real-time predictive instrument was developed and tested by a team of researchers whose work was supported by the Agency for Health Care Policy and Research (HS06208). The description of the instrument and all necessary details to calculate its predictions for any patient with a heart attack appeared in the October 1 Annals of Internal Medicine.

According to AHCPR's Administrator John M. Eisenberg, M.D., physicians currently base thrombolytic therapy decisions mostly on the same kinds of information the instrument uses, but with the new tool, the process will be more systematic, precise, and practical.

Thrombolytic drugs help many patients survive a heart attack if administered in time. However, if the drugs are administered too late, or given to patients unlikely to benefit, they are of little or no help, may cause serious complications, and waste money. Other than timing (AHCPR-funded and other research shows that heart attack victims often delay seeking care), the main reasons more patients do not receive thrombolytic therapy is failure to recognize suitable candidates and fear of complications.

According to the research team leader, Harry P. Selker, M.D., of Boston's New England Medical Center, the new instrument estimates the patients' probabilities (from zero to 100 percent) of suffering cardiac arrest within 48 hours, of dying within 30 days, or of dying within 1 year, if the patient is, or is not, treated with thrombolytic therapy in the emergency room. The instrument also calculates the patients' likelihoods of developing serious complications, such as hemorrhagic stroke or major bleeding, if given thrombolytic therapy.

The study involved researchers from New England Medical Center/Tufts University School of Medicine, Duke University Medical Center, the University of Washington School of Medicine, Boston Universitys Department of Mathematics, Harbor-UCLA Medical Center, and the University of Southern California's Department of Medicine.

AHCPR is currently funding a followup to Dr. Selker's study—a randomized controlled trial in a diverse group of 28 hospitals across the country to test the clinical impact of making this thrombolytic predictive instrument available in emergency rooms.

For details of the current study, see "Patient-specific predictions of outcomes in myocardial infarction for real-time emergency use: A thrombolytic predictive instrument," by Dr. Selker, John L. Griffith, Ph.D., Joni R. Beshansky, R.N., M.P.H., and others, in the October 1, 1997, issue of the Annals of Internal Medicine, 127(7), pp. 538-556.

Return to Contents

Research on Heart Disease

Study reaffirms link between angioplasty outcomes and hospital and physician experience

In the largest study of its kind to date, researchers have once again found that elderly patients undergoing coronary angioplasty are well advised to have the procedure done in hospitals that perform 200 or more angioplasties each year and by physicians who perform 75 or more angioplasties each year. These patients suffer fewer complications and deaths than similar patients who undergo angioplasties at hospitals and by physicians who perform the procedure less often, according to the study, which was supported by the Agency for Health Care Policy and Research (HS08805 and the Ischemic Heart Disease Patient Outcomes Research Team [PORT], HS06503).

In this study, the rate of followup bypass surgery (a marker for failed angioplasty) and/or death was greatest for patients treated by low-volume physicians in low-volume hospitals (4.2 percent and 6.9 percent, respectively) and lowest for patients treated by high-volume physicians in high-volume hospitals (2.5 percent and 4.5 percent, respectively).

The empirical data provided by this study support minimum angioplasty volume standards for physicians and hospitals, notes James G. Jollis, M.D., of Duke University Medical Center. Dr. Jollis and his colleagues examined in-hospital bypass surgery and death after angioplasty for all 1992 Medicare enrollees who had angioplasties, according to 1992 physician and hospital Medicare procedure volume. They found that more than 50 percent of physicians and 25 percent of hospitals performing this procedure in 1992 were unlikely to have met the 1988 American College of Cardiology/American Heart Association (ACC/AHA) minimum volume guidelines of 200 procedures per year for hospitals and 50 procedures per year for surgeons.

Improved outcomes were seen up to threshold values of 75 Medicare cases per physician and 200 Medicare cases per hospital or overall volumes of 150 to 225 cases per physician and 400 to 600 cases per hospital. It is likely that the increased skill of high-volume operators in successfully dilating lesions and performing "bailing out" procedures after abrupt closure of an artery led to the lower rates of followup bypass surgery for these providers. On the other hand, referring physicians may have known which angioplasty providers had better outcomes and selectively referred more patients to such providers. Thus, these providers performed the highest volumes of procedures. In conclusion, Dr. Jollis points out the need for a similar study using more contemporary data to assess the impact of stents and platelet inhibitors on outcomes by volume.

Details are in "Relationship between physician and hospital coronary angioplasty volume and outcome in elderly patients," by Dr. Jollis, M.D., Eric D. Peterson, M.D., M.P.H., Charlotte L. Nelson, M.S., and others, in the June 3, 1997, Circulation 95(11), pp. 2485-2491.

More women than men die in the hospital after a heart attack, but experts disagree about the reasons for this disparity

Women typically develop heart disease a decade later than men, but it continues to be the number one killer of women in the United States. Although men are more apt to have heart attacks (acute myocardial infarctions, AMIs) than women, women are more likely to die from them.

Several studies have pointed to less access for women than men to revascularization procedures and clot-busting (thrombolytic) therapy as an explanation for this difference in heart attack deaths. Other studies have attributed the disparity to women's older age at the time of heart attack and more coexisting medical problems, such as diabetes and hypertension. Two new studies supported by the Agency for Health Care Policy and Research continue this debate and underscore the need for more research on this issue.

The first study (Cardiac Arrhythmia Patient Outcomes Research Team [PORT], HS08362) finds that even after accounting for differences between men and women in cardiac procedures, age, and health factors, women were still 20 percent more likely than men to die in the hospital following heart attack. In contrast, a second study (HS00060 and HS07360) concludes that women's higher mortality rate is due to pulmonary congestion, older age, and greater number of coexisting medical problems than men.

Maynard, C., Every, N.R., Martin, J.S, and others (1997, June). "Association of gender and survival in patients with acute myocardial infarction." Archives of Internal Medicine 157, pp. 1379-1384.

This study of over 12,000 women and men with AMI treated in Seattle area hospitals between 1988 and 1994 shows that women were 20 percent more likely than men to die in the hospital. Women were 8 years older than men; more often had a history of congestive heart failure, hypertension, or diabetes mellitus; but less often had a history of AMI or coronary surgery. Women were less apt than men to receive clot-busting therapy within 6 hours of hospital admission (15 percent vs. 25 percent) and to undergo coronary angiography to diagnose heart problems (46 percent vs. 63 percent), angioplasty (21 percent vs. 29 percent), or bypass surgery (9 percent vs. 13 percent).

Even after adjusting for differences by sex in receipt of cardiac procedures, age, and health factors, women's 20 percent greater risk of in-hospital death remained.

In addition, many of these older women (median age of 73 years) had long delay times from symptom onset to hospital arrival and significant coexisting illnesses that decrease the benefits and increase the risk of complications from thrombolytic therapy or revascularization. Consequently, many of the women probably were not suitable candidates for these cardiac procedures, explain University of Washington researchers who conducted the study. They also found that after adjusting for age, survival after hospital discharge was similar for men and women.

Coronado, B.E., Griffith, J.L., Beshansky, J.R., and Selker, H.P. (1997, June). "Hospital mortality in women and men with acute cardiac ischemia: A prospective multicenter study." Journal of the American College of Cardiology 29, pp. 1490-1496.

The difference in hospital mortality rates for women and men who suffer heart attacks (10 percent for women vs. 7 percent for men) appears to be accounted for by women's higher degree of pulmonary congestion when they arrive at the emergency department (ED), older age, and greater incidence of coexisting medical conditions, according to researchers at the New England Medical Center and Tufts University. Their findings contrast with other studies, including the study described above, where differences between men and women persisted after controlling for these and other clinical factors.

The researchers analyzed data from a prospective multicenter study involving 10,783 older patients (nearly half were women) presenting to the ED with symptoms suggesting acute cardiac ischemia (restricted blood flow to the heart). These included chest pain, dizziness, nausea, vomiting, jaw pain, difficult or labored breathing, chest pressure, upper abdominal pain, and left arm pain.

Women had more hypertension (55 percent vs. 46 percent) and diabetes (23 percent vs. 17 percent) than men but fewer previous AMIs (21 percent vs. 29 percent). Women who presented with an AMI also had more extensive heart damage than men. Age, diabetes, presenting systolic blood pressure, and radiographic evidence of heart failure significantly predicted death from acute cardiac ischemia and heart attack. Once these factors were accounted for, the trend for higher mortality noted in women was no longer present.

Beta-blockers reduce deaths from congestive heart failure

About 40,000 persons in the United States die from congestive heart failure (CHF) each year. Angiotensin-converting enzyme inhibitors significantly reduce death among these patients and as a result have become the standard therapy for this condition. Several other drugs have been tried with inconclusive or less-effective results. However, a recent meta-analysis of several studies shows that beta-blocker therapy reduces the overall odds of death by 31 percent in patients with chronic CHF, with one death prevented per 35 patients treated. This therapy improves survival most for progressive, nonsudden rather than sudden cardiac death, with use of the alpha- and beta-blocking agent carvedilol; the improved survival is unrelated to the cause of heart failure, according to the study, which was supported by the Agency for Health Care Policy and Research (National Research Service Award training grant T32 HS00028).

Beta-blockers reduced nonsudden cardiac death by 42 percent and sudden cardiac death by 16 percent. This difference was most notable in the noncarvedilol trials, which demonstrated a 34 percent decrease in the odds of nonsudden cardiac death and a 4 percent increase in the odds of sudden cardiac death. This suggests that beta-blockers are more likely to prevent or delay the progression of CHF (to nonsudden death) than exert a primary antiarrhythmic effect (to prevent sudden death), note Paul A. Heidenreich, M.D., and Tina T. Lee, M.D., of Stanford University, and Barry M. Massie, M.D., of the University of California, San Francisco.

The researchers conducted a meta-analysis of 17 randomized clinical trials of beta-blockade treatment in 3,039 patients with CHF from January 1995 through February 1997. Based on all 17 reports, beta-blockade therapy significantly reduced mortality (odds ratio, OR of 0.69; 1 is equal odds). A trend toward greater treatment effect was noted for nonsudden cardiac death (OR, 0.58) compared with sudden cardiac death (OR, 0.84). Similar reduced deaths (OR, 0.69) were observed for patients with ischemic and nonischemic cardiomyopathy. The survival benefit was greater for trials of the drug carvedilol (OR, 0.54) than for noncarvedilol drugs (OR, 0.82), but the difference was not significant. The researchers point out, however, that most of the studies followed patients for 1 year or less, so the effect of beta-blockers on long-term survival is less clear.

See "Effect of beta-blockade on mortality in patients with heart failure: A meta-analysis of randomized clinical trials," by Drs. Heidenreich, Lee, and Massie, in the July 1997 Journal of the American College of Cardiology 30, pp. 27-30.

Cardiologists and internists disagree about the risks associated with frequently performed cardiac procedures

Cardiac procedures such as cardiac catheterization, angioplasty, and bypass surgery carry some risks of serious complications and death. In general, cardiologists see these procedures as less risky than do internists, but their judgments are not always more accurate, concludes a study supported in part by the Agency for Health Care Policy and Research (HS06274).

Under managed care arrangements, conflicts could arise between generalist gatekeepers, who believe the invasive cardiac procedures are relatively more hazardous, and cardiologists, who believe they are less risky. These conflicts could be difficult to settle because the evidence available in the medical literature is insufficient to determine which group of physicians is right, according to Roy M. Poses, M.D., and Wally R. Smith, M.D., of the Medical College of Virginia, and their colleagues. The researchers based their findings on a survey of senior staff and physician-trainees at two hospitals affiliated with a U.S. medical school and two hospitals associated with a U.K. medical school.

Based on risk assessment of four cardiac procedures, cardiologists at the four hospitals systematically judged the risks to be much lower than hospital internists. Cardiologists and internists each judged the risks accurately about half the time. Cardiologists were more accurate in their judgments of the rates of serious problems (morbidity) and death due to cardiac catheterizations, but internists were more accurate in their judgments of rates of morbidity due to Swan-Ganz catheterization and angioplasty. There were no clear differences between cardiologists' and internists judgments of death due to Swan-Ganz catheterization, angioplasty, or bypass surgery or morbidity due to bypass surgery.

Cardiologists' personal experiences with these procedures may encompass many more good outcomes than bad ones. Internists, however, may only become personally involved in such procedures when they are called about bad outcomes. On the other hand, physicians may look more favorably on the treatments they use than on treatments used by other physicians. Finally, disagreements about the risks of procedures may arise from a paucity of published data on the risks of cardiac catheterization or from an abundance of confusing, contradictory data on angioplasty and bypass surgery.

See "Physicians' judgments of the risks of cardiac procedures: Differences between cardiologists and other internists," by Dr. Poses, Donna K. McClish, Ph.D., Dr. Smith, and others, in Medical Care 35(6), pp. 603-617.

Return to Contents

Medical Effectiveness/Outcomes Research

Low Birthweight PORT publishes recent findings

Conditions ranging from maternal low weight and substance abuse to vaginal infections and sexually transmitted diseases have been linked to low birthweight (LBW) and other neonatal problems. The goals of the Patient Outcomes Research Team (PORT) on the Prevention of Low Birthweight in Minority and High-Risk Women are to better define the risk factors associated with LBW, determine its causes, and evaluate the effectiveness of strategies aimed at the prevention of LBW. The PORT is led by Robert L. Goldenberg, M.D., of the University of Alabama at Birmingham and supported by the Agency for Health Care Policy and Research (PORT contract 290-92-0055).

The PORT researchers recently published the following three articles. One, the cost of caring for infants who contract congenital syphilis from mothers with syphilis is more than three times that of infants without this problem. Two, high blood levels of alkaline phosphatase at 26 weeks' gestation and changes in activities of this enzyme between 19 and 26 weeks' gestation are both associated with a three-fold increase in the risk for preterm delivery. And three, vaginal irrigation with the medical disinfectant chlorhexidine during active labor or before planned cesarean section is not effective in preventing maternal infection shortly before or after childbirth. The articles are summarized below.

Bateman, D.A., Phibbs, C.S., Joyce, T., and Heagarty, M.C. (1997, May). "The hospital cost of congenital syphilis." The Journal of Pediatrics 130(5), pp. 752-758.

Congenital syphilis has risen sharply in the United States, reaching a peak in 1989 to 1991, with most cases occurring in urban minority groups. Infants contract congenital syphilis from their mothers, either prenatally or at birth. The Centers for Disease Control and Prevention (CDC) guidelines for treatment of congenital syphilis require that newborns suspected of having this problem be given parenteral penicillin therapy for at least 10 days following delivery, usually while hospitalized. The unadjusted mean cost of caring for these infants ($11,031) is more than three times that of caring for infants without syphilis, according to this study.

The researchers used Medicaid reimbursement data to compare the cost of caring for 114 newborn infants treated for confirmed or presumptive congenital syphilis with the cost of caring for 2,906 infants without syphilis at Harlem Hospital in New York City in 1989.

After controlling for maternal behaviors and newborn risk factors that might influence length of neonatal hospitalization, infants with syphilis had an additional hospital stay of 7.5 days and an additional cost of $4,691 above mean study population values (7.13 days, $3,473). A maternal history of injection drug use added another 7.7 days at a cost of $3,946. Based on the number of reported cases of congenital syphilis from 1991 to 1994, the average national cost per year of treating these infants is about $18.4 million (1995 dollars).

Congenital syphilis is a preventable disease. Preventing cases of maternal syphilis and treating pregnant women for syphilis would reduce the number of congenital cases needing treatment. The additional cost of treating syphilis during pregnancy—for repeated blood tests, one or two additional doctor visits, and penicillin—would add several hundred dollars to the cost of a routine prenatal care program. Given the higher rate of intrauterine cocaine exposure among neonates with congenital syphilis in this study (70 percent), as well as the cost of this exposure, programs that combine prenatal care with maternal treatment for substance use and other high-risk behaviors may be more effective and yield additional savings, conclude the researchers.

Goldenberg, R.L., Tamura, T., DuBard, M., and others (1997). "Plasma alkaline phosphatase and pregnancy outcome." Journal of Maternal-Fetal Medicine 6, pp. 140-145.

Alkaline phosphatase is an enzyme found in tissues throughout the body, including the placenta, and increases in plasma activity occur as pregnancy approaches term. Several studies have suggested a relationship between elevated alkaline phosphatase activity and low birthweight, preeclampsia, prematurity, and placental insufficiency. This study finds that high plasma alkaline phosphatase activity at about 26 weeks' gestation and changes in alkaline phosphatase activity between 19 and 26 weeks' gestation are both associated with about a three-fold increased risk for preterm delivery. This finding remains significant after adjusting for a number of maternal characteristics and for plasma copper, folate, and albumin, substances that are associated with alkaline phosphatase activity.

Most significant risk factors for preterm delivery predict an odds ratio (OR) of 2.0 or less. This study found that alkaline phosphatase activity at 26 weeks' gestation had an OR of 5.0 for delivery at 32 weeks or earlier, suggesting that it is a relatively powerful predictor of early preterm delivery. The researchers hypothesize that infection of the chorioamnion, the outermost embryonic membrane that gives rise to the placenta, or disruption of the interface between the embryonic membrane and uterine lining by some other mechanism, may increase the production or release of placental alkaline phosphatase into the circulation. The researchers caution, however, that at this time there is insufficient information to recommend the routine use of tests for measuring plasma alkaline phosphatase for the prediction of preterm delivery and low birthweight.

Rouse, D.J., Hauth, J.C., Andrews, W.W., and others (1997). "Chlorhexidine vaginal irrigation for the prevention of peripartal infection: A placebo-controlled randomized clinical trial." American Journal of Obstetrics & Gynecology 176, pp. 617-622.

Chlorhexidine is a widely used medical disinfectant with antimicrobial activity against a broad spectrum of vaginal bacteria. It has been shown to reduce neonatal morbidity associated with group B streptococci. This study tested whether a single vaginal irrigation with dilute chlorhexidine solution administered during active labor or before cesarean delivery would be effective in preventing maternal infection during the last month of pregnancy or the first 3 months after delivery. The researchers administered chlorhexidine during active labor or before planned cesarean delivery to 508 women and sterile water placebo to 516 women.

Rates of infection (intrapartum chorioamnionitis and postpartum endometritis) did not differ significantly between the groups, with 10 percent of women receiving chlorhexidine having infections versus 13 percent of women who received the sterile water. Neonatal outcomes, including sepsis rates of 0.4 percent, were equivalent for both groups. Thus, vaginal irrigation with dilute chlorhexidine solution about 4.5 hours before delivery did not prevent maternal peripartal infection. Either the solution lacked activity against the bacteria responsible for the infections, it was not given early enough or should have been repeated, or the irrigant was not adequately dispersed throughout the vagina for other reasons, conclude the researchers.

Most people take more than 30 days following pneumonia to be symptom-free

For most persons with community-acquired pneumonia (CAP), it takes at least 30 days for the symptoms of cough, chest pain, labored breathing (dyspnea), and fatigue to resolve. When symptoms take longer to disappear, individuals generally make additional visits to the doctor, according to a study by the Pneumonia Patient Outcomes Research Team (PORT), which is supported by the Agency for Health Care Policy and Research (HS06468) and led by Wishwa N. Kapoor, M.D., M.P.H., of the University of Pittsburgh.

The researchers studied 576 adults with clinical and x-ray evidence of pneumonia who were judged to be at low risk of dying. They documented the presence and severity of five symptoms (cough, fatigue, dyspnea, sputum, and chest pain) through questionnaires administered to the patients at diagnosis and at 1 week, 1 month, and 3 months later. The researchers also determined the reason for and number of outpatient physician visits during that time.

Physical health had declined significantly at diagnosis but improved during the subsequent followup. With the exception of pleuritic chest pain, symptoms continued to be a problem at 30 days, and some symptoms continued to be bothersome at 90 days. For instance, fatigue was reported by 51 percent of patients at day 30 with 20 percent still reporting moderate to severe fatigue at day 90; 32 percent of patients reported cough and 27 percent reported sputum 90 days after diagnosis.

Patients with elevated symptom scores at 1 week or 1 month after diagnosis were significantly more apt to report pneumonia-related outpatient care visits 1 month and 3 months later. For instance, 43 percent of patients with the fewest symptoms a week after diagnosis subsequently reported pneumonia-related outpatient visits by day 30 compared with 72 percent of patients with the most symptoms. This relationship remained significant even after adjusting for the patients age, site of care, coexisting conditions, and severity of illness. Physicians should inform patients with pneumonia of the possibility that their symptoms may not resolve for quite some time. It is possible that if physicians better informed patients, the rate of pneumonia-related outpatient visits might decrease, conclude the researchers.

See "Measuring symptomatic and functional recovery in patients with community-acquired pneumonia," by Joshua P. Metlay, M.D., Ph.D., Michael J. Fine, M.D., M.Sc., Richard Schulz, Ph.D., and others in the July 1997 Journal of General Internal Medicine 12, pp. 423-430.

Return to Contents
Proceed to Next Section

The information on this page is archived and provided for reference purposes only.


AHRQ Advancing Excellence in Health Care