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

Medical Effectiveness/Outcomes Research

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 It was current when produced and may now be outdated. Archive material is no longer maintained, and some links may not work.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.

Recent findings from the Cardiac Arrhythmia PORT

The following two studies were reported recently by members of the Cardiac Arrhythmia Management Patient Outcomes Research Team (PORT-II), which is led by Mark A. Hlatky, M.D., of Stanford University and supported by the Agency for Health Care Policy and Research (HS08362). The goal of the PORT-II researchers is to develop a comprehensive decision model for use in screening and treating patients at risk for sudden cardiac death due to ventricular arrhythmias.

Sim, I., Gupta, M., McDonald K., and others. (1995). "A meta-analysis of randomized trials comparing coronary artery bypass grafting with percutaneous transluminal coronary angioplasty in multivessel coronary artery disease." American Journal of Cardiology 76, pp. 1025-1029.

Heart attack patients who have more than one coronary artery blocked are as likely to die or have a nonfatal heart attack within 1 to 3 years whether they undergo percutaneous transluminal coronary angioplasty (PTCA) or coronary artery bypass graft (CABG) surgery. These revascularization procedures open up blocked arteries to reestablish blood flow (reperfusion). However, CABG provides better relief of angina (severe heart pain) and leads to fewer repeat revascularization procedures, according to Cardiac Arrhythmia PORT-II investigators.

They performed a meta-analysis of all randomized trials published from 1985 to 1995 that directly compared CABG surgery and PTCA in patients with multivessel coronary disease. Results showed that the overall risk of death and nonfatal heart attack did not differ over a 1- to 3-year followup period. Patients randomized to CABG tended to have a slightly higher risk of death or heart attack during hospitalization but a lower risk in subsequent followup.

CABG patients were much less likely than PTCA patients to undergo either a subsequent CABG (1 percent vs. 19 percent) or PTCA (6 percent vs. 23 percent) and were more likely to be angina-free (81 percent vs. 73 percent). The researchers point out that published data do not provide sufficient detail to allow analysis by number of diseased vessels. However, observational data suggest that PTCA yields better outcomes than CABG for 2-vessel disease but poorer outcomes for 3-vessel disease.

Weaver, W.D., Parsons, L., and Every, N. (1995). "Primary coronary angioplasty in hospitals with and without surgery backup." The Journal of Invasive Cardiology 7 (Suppl F), pp. 34F-39F.

Use of coronary angioplasty as the first approach to treating heart attack patients is increasing. Sometimes patients who undergo angioplasty subsequently need coronary artery bypass graft surgery, if the angioplasty fails to open up clogged arteries. Nevertheless, angioplasty is being performed more often at hospitals that have freestanding cardiac catheterization laboratories but no backup cardiac surgery facilities.

Yet patients undergoing angioplasty at these hospitals have acceptable outcomes compared with patients undergoing the procedure at hospitals with cardiac surgery facilities. Patient age, a history of prior heart attack, and ECG evidence of anterior ST elevation—but not the availability of on-site surgical backup—are associated with patient outcomes, according to the Cardiac Arrhythmia PORT-II investigators.

As part of the Myocardial Infarction Triage and Intervention (MITI) project—a registry of all admissions to Kings County, WA, hospitals for acute cardiac disease—they compared the outcomes of 470 persons who received coronary angioplasty as their first treatment for heart attack in hospitals without on-site coronary surgery capability with those of similar patients treated in hospitals with on-site surgery. The researchers found no differences in procedural success rates or initial and long-term mortality rates in patients treated in the two types of hospitals. The mortality rate at discharge was 7 percent for this procedure in both types of facilities. These data suggest that with appropriate patient selection, trained operators, and provision for hospital transfer, primary coronary angioplasty can be successfully accomplished in centers without on-site surgery capability.

Recurrent cardiac ischemia after TT remains a frequent, serious, and costly problem

The early use of thrombolytic therapy (TT) opens up clogged coronary arteries in patients with acute myocardial infarction (AMI), but it does not remove the clot-producing (thrombogenic) stimulus, which can cause recurrent ischemia (reduced blood flow). A new study, supported in part by the Agency for Health Care Policy and Research (HS05635), shows that recurrent myocardial ischemia after TT remains a frequent and expensive clinical problem, and it is difficult to predict which patients will develop this serious condition.

Researchers from the Fundacion Favaloro, Buenos Aires, Argentina; Duke University Medical Center; the Cleveland Clinic Foundation; and other U.S. sites, examined patients enrolled in the Thrombolysis and Angioplasty in Myocardial Infarction (TAMI) studies at five centers; 552 patients were treated with tissue plasminogen activator (t-PA), 293 were treated with urokinase, and 385 received both thrombolytic agents.

Results showed that recurrent ischemia alone occurred in 18 percent of patients, and both recurrent ischemia and reinfarction (newly destroyed heart tissue due to lack of blood flow) occurred in 3.4 percent of patients; 78 percent of patients experienced neither cardiac problem. Women, older patients, and those with diabetes or hyperlipidemia had more reinfarction and recurrent ischemia. Patients with reinfarction had significantly higher rates of in-hospital death (21 percent) and heart failure (50 percent) than patients with recurrent ischemia alone (11 percent and 31 percent, respectively) or no ischemic events (4 percent and 17 percent). As expected, median in-hospital costs were highest in patients with reinfarction ($26,802), intermediate for those with recurrent ischemia alone ($18,422), and lowest in patients with neither event ($15,623).

More details are in "Frequency, significance, and cost of recurrent ischemia after thrombolytic therapy for acute myocardial infarction," by Alejandro Barbagelata, M.D., Christopher B. Granger, M.D., Eric J. Topol, M.D., and others, in the November 15, 1995 American Journal of Cardiology 76, pp. 1007-1013.

Low Birthweight PORT examines use of prenatal testing and corticosteroids in prevention of preterm birth

Potential problems with the use of prenatal testing at progressively earlier gestational ages, prevention of preterm deliveries in pregnant women with bacterial vaginosis, and controversies surrounding the use of prenatal corticosteroids are discussed in three recent articles, summarized here, by members of the Patient Outcomes Research Team (PORT) on Low Birthweight in Minority and High-Risk Women. Supported by the Agency for Health Care Policy and Research (contract 282-92-0055), the PORT examines ways to prevent low birthweight and improve the outcomes of low-birthweight infants. The Low Birthweight PORT is led by Robert L. Goldenberg, M.D., of the University of Alabama at Birmingham.

Rouse, D.J., Owen, J., Goldenberg, R.L., and Cliver, S.P. (1995, November). "Determinants of the optimal time in gestation to initiate antenatal fetal testing: A decision-analytic approach." American Journal of Obstetrics and Gynecology 173(5), pp. 1357-1363.

Antenatal fetal testing is widely used in clinical obstetric practice as a method of determining fetal well-being. However, according to this decision analysis by the Low Birthweight PORT, the use of fetal testing at progressively earlier gestational ages is particularly dangerous because the tests do not have perfect specificity, resulting in high false-positive rates. For example, acting on a false-positive test by effecting delivery at 28 weeks' gestation will predictably have more harmful consequences for the infant than at 34 weeks. The PORT researchers used decision analytic techniques to model the most important determinants of the optimal timing to begin antenatal fetal testing. Because pregestational insulin-dependent diabetes mellitus is one of the more common maternal indications for these tests, the analysis uses a contemporary estimate for the risk of stillbirth in these pregnancies as the baseline fetal death risk. The results of the analysis showed that even with the increased risk of fetal death associated with maternal conditions such as insulin-dependent diabetes, very high test specificity (more than 99 percent) is required to recommend testing before 30 weeks. Because of the lower specificity (higher false-positive rates) of these tests, neonatal deaths will increase due to an increase in unnecessary and earlier preterm deliveries. The analysis also showed that the increased risk of neonatal deaths is not offset by a corresponding decrease in fetal deaths. Under the baseline assumptions of this model, the researchers conclude that 34 weeks is a more optimal time to initiate antenatal testing than 30 weeks because it results in more neonatal survivors.

Hauth, J.C., Goldenberg, R.L., Andrews, W.W., and others. (1995, December). "Reduced incidence of preterm delivery with metronidazole and erythromycin in women with bacterial vaginosis." The New England Journal of Medicine 333, pp. 1732-1736.

Treatment with metronidazole plus erythromycin (M+E) reduces rates of premature delivery in pregnant women with bacterial vaginosis who also are at risk of preterm delivery, according to this study by Low Birthweight PORT investigators. The study recruited 624 pregnant women identified for one of two risk factors for preterm delivery: previous spontaneous preterm delivery or prepregnancy weight of less than 50 kg (approximately 111 lb). These women were subsequently screened for several lower genital tract infections, including asymptomatic bacterial vaginosis (BV), and randomized (2:1) to receive M+E versus placebo at 23 weeks' gestation. Of the 624 women randomized, 258 had asymptomatic bacterial vaginosis at baseline examination, 358 were negative for BV, and 8 were lost to followup. The initial analysis indicated that, overall, the treatment group receiving metronidazole plus erythromycin had a preterm delivery rate of 26 percent compared with 36 percent in the placebo group. However, after further subanalysis, a lower rate of prematurity in the M+E versus placebo-treated women was observed only in the 258 subjects with BV at baseline examination: 31 percent versus 49 percent. There was virtually no difference in the rate of prematurity between the study treatment and placebo groups in the 358 women who did not have BV at baseline examination (22 percent vs. 25 percent, respectively). The lower rate of preterm delivery in the 258 BV-positive women who received the study treatment versus placebo was observed both in the previous preterm delivery risk group (39 percent treatment vs. 57 percent placebo) and in the group of women with a prepregnancy weight of less than 50 kg (14 percent treatment vs. 33 percent placebo).

Gardner, M.O., and Goldenberg, R.L. (1995, December). "The clinical use of antenatal corticosteroids." Clinical Obstetrics and Gynecology 38(4), pp. 746-754.

Prenatal use of corticosteroids has been shown to have many beneficial effects for the preterm infant, including reducing the incidence and severity of respiratory distress syndrome, intraventricular hemorrhage, necrotizing enterocolitis (inflammation of the small intestine and colon), and neonatal death. Nevertheless, the available estimates of use suggest that only 20 percent of eligible mothers who deliver before 34 weeks receive corticosteroids. Concern about the disparity between scientific evidence of benefit and clinical practice led the National Institutes of Health (NIH) to convene a consensus conference in February 1994. In this paper, members of the Low Birthweight PORT examine areas of controversy, such as corticosteroid treatment in pregnancies with premature rupture of membranes. They also review practice patterns in the United States, including surveys of physician attitudes toward corticosteroid use, and summarize the NIH consensus conference recommendations.

Surgery found effective in reducing seizures and medications needed to control intractable epilepsy

Adults and adolescents with intractable epilepsy who are treated surgically have better seizure control with less antiepileptic medication than patients who do not undergo surgery, according to a study supported in part by the Agency for Health Care Policy and Research (HS06856). Surgical treatment of epilepsy requires a substantial series of presurgical diagnostic tests to identify the brain region producing the epileptic attacks and determine if surgical removal of the affected tissue is possible without damaging other brain functions.

No randomized trials have been conducted to determine the effectiveness of surgical treatment for epilepsy. This is the first prospective, multivariate analysis of outcomes comparing epilepsy patients who were treated surgically with those who were not.

Researchers, led by Barbara G. Vickrey, M.D., of the University of California, Los Angeles, and the RAND Corporation, evaluated the impact of epilepsy surgery on seizures, medication use, employment, and quality of life in 248 adults and adolescents with hard-to-control epilepsy, who underwent diagnostic evaluation for surgery at the UCLA medical center between 1974 and 1990. Those in whom an epileptogenic focus could not be found did not undergo surgery and were treated medically. After undergoing screening, 202 patients qualified for and underwent surgery.

Results showed that nearly 6 years later, on average, over 80 percent of non-surgery patients, but only 25 percent of surgery patients, were still having more than one seizure per month at followup. Nearly 60 percent of surgery patients were either completely seizure-free or having only auras or one seizure, compared with 11 percent of non-surgery patients.

On average, surgery patients were taking almost one fewer antiepileptic drug at followup than at study enrollment (1.4 vs. 2.0 antiepileptic medications). Although the surgery group showed no advantage at followup in employment status or self-reported quality of life according to prospectively collected measures, quality-of-life scores were higher with surgery on 5 of 11 scales of a measure designed for epilepsy patients but administered only at followup. These outcomes need to be assessed in larger prospective studies, according to the researchers.

For more details, see "Outcomes in 248 patients who had diagnostic evaluations for epilepsy surgery," by Dr. Vickrey, Ron D. Hays, Ph.D., Rebecca Rausch, Ph.D., and others, in the December 2, 1995 issue of Lancet 346, pp. 1445-1449.

High mortality found for patients with community-acquired pneumonia

Nearly 14 percent of patients with community-acquired pneumonia (CAP) die, according to a meta-analysis of 122 studies involving more than 33,000 patients conducted by the Pneumonia Patient Outcomes Research Team (PORT). The Pneumonia PORT is led by Wishwa N. Kapoor, M.D., M.P.H., of the University of Pittsburgh, and supported by the Agency for Health Care Policy and Research (HS06468). Mortality ranged from 5 percent for studies that covered both ambulatory and hospitalized patients and 14 percent for studies that included only hospitalized patients to 31 percent for nursing home residents and 37 percent for intensive care unit patients.

Men were more likely to die than women, as were patients with hypothermia, systolic hypotension, tachypnea (a condition marked by quick, shallow breathing), diabetes mellitus, neoplastic disease, neurologic disease, bacteremia, leukopenia, and pulmonary infiltrate found in more than one lobe of the lungs on x-ray. Mortality was strongly associated with the cause of pneumonia. In the case of viral pneumonia, mortality was lowest for patients with influenza B (0 percent) and adenovirus pneumonia (0 percent) and ranged from 5 to 9 percent for the remaining viral etiologies. For bacterial pneumonia, mortality was highest for patients with Pseudomonas aeruginosa (61.1 percent), Klebsiella species (35.7 percent), Escherichia coli (35.3 percent), and Staphylococcus aureus (31.8 percent).

Patients suspected of having one of these high-risk, gram-negative rod or staphylococcal bacterial infections should be considered appropriate candidates for inpatient care and broad-spectrum antibiotic therapy, according to Michael J. Fine, M.D., M.Sc., lead author of this study, and other PORT members. They recommend more widespread use of pneumococcal vaccination in at-risk persons because of the high prevalence and mortality associated with pneumococcal pneumonia and the increased resistance of this bacteria to penicillin.

Details are in "Prognosis and outcomes of patients with community-acquired pneumonia," by Dr. Fine, Melanie A. Smith, M.P.I.A., Catherine A. Carson, Ph.D., and others, in the January 10, 1996 Journal of the American Medical Association 275(2), pp. 134-141.

Return to Contents
Proceed to Next Article

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


AHRQ Advancing Excellence in Health Care