Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness (ESCAPE)

BACKGROUND:

Congestive heart failure constitutes one of the major categories of morbidity, particularly in the elderly, and is responsible for the utilization of significant resources, including a large number of hospitalizations (estimated at 800,000 to 2.3 million per year) and related health care costs. In 1993, 4.7 million Americans carried the diagnosis of CHF, and about 400,000 new cases are diagnosed every year. It is estimated that there are 800,000 to 1.2 million CHF patients with NYHA class III-IV symptoms. The diagnosis and treatment of CHF and its episodes of decompensation can be facilitated by pulmonary artery catheter use, but its efficacy is highly dependent on the physician's experience in the treatment of such patients. Recently, concern has been raised about the benefits of PAC, also known as Swan-Ganz catheterization.

In response to concerns about the benefits and safety of PAC, the NHLBI and the Food and Drug Administration (FDA) co-sponsored a workshop in August 1997 entitiled "Pulmonary Artery Catheterization and Clinical Outcomes (PACCO)". The purpose of the workshop was to provide an objective assessment of the state-of-the-science of PAC and its various uses. Experts in critical care, pulmonary medicine, cardiovascular medicine, surgery, pediatric cardiology, nursing, biostatistics, ethics, and medical economics identified several important clinical areas as priorities for clinical trials; persistent/refractory congestive heart failure (NYHA class IV) received top priority to test whether a PAC-directed treatment strategy achieves a better and less costly long-term outcome compared to a non-invasive treatment strategy. In addition to providing data on outcome, hospital utilization, and costs, the study would have additional benefits. For example, it would provide a clear benchmark for testing other technologies, now in developmental stages, that could supplant PAC in the future. Finally, the study would provide the basis for developing competency requirements for physicians, nurses, and others who insert and use pulmonary artery catheters, which was a strong recommendation of the PACCO Workshop.

DESIGN NARRATIVE:

Patients were randomized into two treatment arms: A total of 218 patients were assigned into a treatment strategy without PAC hemodynamic monitoring (Usual Group) and 215 patients into a treatment strategy guided by PAC (Hemodynamic Group). The primary endpoint was the combined endpoint of rehospitalization and death.

For all patients randomized, therapy was tailored to the ultimate goal of discharge on an oral medical regimen to provide better relief of CHF symptoms, to reduce filling pressures and to maintain adequate perfusion. These goals were the same for both groups, but in the Usual Group therapy was adjusted according to clinical assessment alone, while in the PAC-directed group, actual measurement of hemodynamics was used to supplement clinical assessment. The oral medical regimen for discharge was designed by the attending heart failure physician based on the standard available medications of angiotensin converting enzyme inhibitors (ACEI), nitrates, hydralazine, furosemide and other diuretic agents. Intravenous diuretics were used when major diuresis was desired with subsequent change to oral diuretics at least 48 hours before discharge. Physicians could also use low dose infusions of dopamine or dobutamine to facilitate diuresis. Intravenous nitroprusside could be used in either group.

In the Usual Group, therapy was tailored to the following goals: a) absence of evidence of elevated intracardiac filling pressures; relief of orthopnea; relief of abdominal discomfort attributed to hepatosplenchnic congestion; resolution of peripheral edema, ascites, and rales; reduction of jugular venous distension to 5 cm or less above sternal angle; and b) adequate peripheral perfusion; warm extremities, if they could be achieved, and pulse pressure 25 percent or more , or, if less, the greatest which could be achieved. Either during or after hemodynamic monitoring, therapy could be adjusted to achieve the clinical goals above and further adjusted, if necessary to maintain adequate blood pressure and renal function for all patients. Patients for whom therapy was adjusted without hemodynamic monitoring could crossover for hemodynamic monitoring at any time they reached certain criteria. The criteria for crossover included: need to add intravenous inotropic agents above 3 mcg/kg/minute to prevent symptomatic hypotension; repeated inability to discontinue low-dose inotropic agents; or renal insufficiency.

In the PAC-directed hemodynamic arm, therapy was adjusted to achieve specific hemodynamic goals in addition to the clinical goals in the Usual Group. These goals could include: pulmonary capillary wedge pressure (PCWP) 15 mm Hg or less; right atrial pressure 8mm Hg or less unless PCWP is 15 mm Hg or less; maintenance of systolic blood pressure of 80 or above, except for occasional transient decreases to lower levels after oral medications.

Regardless of the goals by which therapy was adjusted in the hospital, the following conditions were met prior to discharge: 24 hours on oral medications alone without major medication change except for anticoagulation, and 48 hours after discontinuation of any intravenous inotropic medications; stable fluid balance; patient education; and home services and follow-up appointment.

Post-discharge management was the same for both groups, according to standard practice. Patients were followed for adjustment of CHF medications in their heart failure clinics at one to two weeks, four weeks, three months, and six months and more often if clinically indicated. There was a minimum of six months of follow-up.

Data collection included: blood pressure and weight at each visit; electrocardiogram at pre-randomization visit and at each hospitalization; serum electrolytes; echocardiogram at randomization, at hospitalization discharge, and at six month follow-up; exercise stress tests during index hospitalization and six month follow-up; natriuretic hormone profile at randomization, discharge, and six month follow-up; medical history, endpoints, side effects, hospitalizations and quality of life; and compliance with protocol.

The study consisted of a Clinical Coordinating Center which included a network of 26 clinical units, and an echocardiographic core laboratory. The Protocol developed extended for six months, patient recruitment extended for 48 months, follow-up for six months, and data analysis for six months.

The NHLBI has awarded R01HL67691 in April 2001 as an ancillary study. The study entitled "ESCAPE Mechanistic Substudies" has its own site on this database.