ESCAPE

Published

Binanay C, et al. "Evaluation study of congestive heart failure and pulmonary artery catheterization effectiveness". JAMA. 2005. 294(13):1625-1633.
PubMed • PDF

Clinical Question

In patients hospitalized for acute decompensated heart failure, does placement of a pulmonary artery catheter and its resultant data improve survival when when compared to clinical assessment alone?

Bottom Line

Among patients admitted with acute decompensated heart failure, pulmonary artery catheter-guided therapy does not improve survival and increases adverse events compared to clinical assessment-guided therapy.

Major Points

Pulmonary artery catheters (PAC), including the Swan-Ganz catheter developed in the 1960s-1970s, allow direct measurement of central hemodynamics at the bedside. This has been widely regarded as useful in patients with acute decompensated heart failure (ADHF) as it allows tailored therapy based upon a patient's response to diuretics, vasopressors, and inotropes. Their use was common in ADHF management at the turn of the millennium despite a lack of randomized trials demonstrating effectiveness.

The 2005 Evaluation Study of Congestive Heart Failure and Pulmonary Artery Catheterization Effectiveness (ESCAPE) randomized 433 patients hospitalized for ADHF to therapy guided by clinical assessment with PAC or clinical assessment alone. At 6 months, there was no difference in days alive out of the hospital or mortality. However, the PAC group had improvement in quality of life and impact of disease in the first few months after discharge. This was offset by a much higher rate of adverse events. The lack of mortality benefit was supported in a concurrently published meta-analysis.

Similarly, the PAC-MAN trial, also published in 2005, did not find a benefit in PAC use in ICU patients, though only a small percentage of these patients had HF. However, the subsequent FACTT trial (2006) demonstrated improved ventilator-free days in patients with ARDS with a PAC or CVC-directed conservative fluid management strategy. This had no affect on mortality.

Because of the outcomes of ESCAPE, some experts have deemed that management of ADHF should not routinely involve PAC-directed therapy. However, its use still may be warranted with failure of routine medical therapy, in assessment of volume and perfusion status for therapy titration as with worsening hypotension or renal dysfunction, in minimizing the risk of arrhythmias or hypotension by guiding optimal dosing of inotropes of vasoactive medications, or in demonstrating benefit from chronic outpatient infusion of inotropes.

Guidelines

AHA/ACCF Heart Failure Guidelines (2013, adapted)

• Invasive hemodynamic monitoring with a PA catheter should be used to guide therapy in patients with respiratory distress or clinical evidence of impaired perfusion when adequacy or excess of intracardiac filling pressures can't be determined by clinical assessment (class I, level C)
• Invasive hemodynamic monitoring can be useful in in acute HF with persistent symptoms despite seemingly appropriate adjustments of standard therapies if any of the following is true (class IIa, level C):
  • Uncertain fluid status, perfusion, systemic vascular resistance, or pulmonary vascular resistance
  • Systolic pressure that remains low or is associated with symptoms despite initial therapy
  • Worsening renal function with therapy
  • Requiring IV vasoactive agents
  • May need mechanical cardiac support or transplant
• Routine invasive monitoring of hemodynamics is not recommended in normotensive patients with acute decompensated HF responding symptomatically to diuretics and vasodilators (class III, level B)

Design

• Multicenter, open-label, randomized comparative trial
• N=433
  • PAC (n=215)
  • Clinical assessment only (n=218)
• Setting: 26 heart failure centers in the US and Canada
• Enrollment: 2000-2003 (stopped early for complications and lack of effect)
• Follow-up: 6 months
• Analysis: Intention-to-treat
• Primary outcome: Days alive out of hospital at six months

Population

Inclusion Criteria

• Severe symptomatic HF as defined by ≥1 of the following:
  • HF hospitalization in the prior year
  • ED visit
  • Treatment with furosemide ≥160 mg/day or equivalent therapy
• ≥3 months of symptoms while on an ACE-inhibitor and diuretics
• LVEF ≤30%
• SBP ≤125 mmHg
• ≥1 sign and ≥1 symptom of congestion

Exclusion Criteria

• Creatinine ≥3.5 mg/dL
• Dobutamine or dopamine ≥3 ug/kg/min or any milrinone on the current hospitalization

Baseline Characteristics

From the PAC group

• Demographics: Age 56 years, male 74%, white race 58%
• HF data: Ischemic 51%, nonischemic 49%, LVEF 19%, 6MWT 390 ft, MLHF score 74
• Baseline health data: HR: 83 BPM, SBP 106 mmHg
• Baseline labs: Na 137 mEq/L, creatinine 1.5 mg/dL, BNP 974 pg/mmol, peak VO₂ 10.2

Interventions

Randomization to clinician-directed therapy guided by clinical assessment with PAC measurements (goals: PCWP 15 mmHg and RA pressure 8 mmHg) or by clinical assessment alone. Clinicians were encouraged to follow national guidelines. Ionotrope use was discouraged.

Outcomes

Comparisons are PAC vs. clinical assessment only.

Primary Outcome

Days alive out of hospital at six months

LVAD/transplant included as mortality: 133 vs. 135 days (HR 1.00; 95% CI 0.92-1.21; P=0.99)

LVAD/transplant included as healthy: 141 vs. 143 days (HR 0.99; 95% CI 0.82-1.21; P=0.95)

Secondary Outcomes

180-day mortality

20% vs. 17% (OR 1.26; 95% CI 0.78-2.03; P=0.35)

This outcome did not change with stratification for inotrope and vasodilator use.

Length of hospital stay

8.7 vs. 8.3 days (HR 1.04; 95% CI 0.78-2.03; P=0.35)

Mean change at discharge

Weight: -4.0 vs. -3.4 kg (P=NS)

SBP: -4 vs. -4 mmHg (P=NS)

JVP: -45% vs. -42% (P=NS)

Creatinine: 0 vs. 0.1 (P<0.05)

Global symptom score: +25 vs. +24 (P=NS)

Orthopnea, 0-4 scale: -1.4 vs. -1.2 (P=NS)

Data visualization for the following two outcomes can be found on Figure 4 on page 1631.

Time trade-off score

A quality of life (QOL) grading system, higher indicating better QOL.

PAC better in PAC group at months 1-6 (P values ranging 0.001-0.02)

MLHF score

A grading system measuring the physical, emotional, mental, and social impacts of heart failure.

1 month: PAC better (statistically significant, see Figure 4)

2 months: No difference between the groups

Subgroup Analysis

There was no difference in the primary outcome for age, sex, race, cardiac index, or site enrollment.

Adverse Events

Only for in-hospital events.

≥1 event

21.9% vs. 11.5% (P=0.04)

AICD firing

2.3% vs. 0.5% (P=0.08)

PAC infection

1.9% vs. 0% (P=0.03)

Cardiogenic shock

0.5% vs. 0.9% (P=0.12)

Ischemia or angina

4.2% vs. 1.8% (P=0.13)

MI

0.0% vs. 0.5% (P=0.75)

Criticisms

• Not blinded
• Only patients with clinical equipose and, therefore, only patients in whom the clinicians felt comfortable managing their ADHF with or without a PAC
• Exclusion of the most critically ill and those being evaluated for assist devices or urgent heart transplant
• Did not evaluate PAC-guided analysis of pulmonary hypertension reversibility as a component of heart transplant workup
• As the clinicians had access to CVP measurements, use of inotropes in the PAC group (a deleterious intervention explicitly excluded from the interventions) may have worsened the group's outcomes
• Adherence was difficult to ensure
• Only clinicians experienced in the management of HF participated
• No defined treatment strategies
• Unclear if non-PAC invasive monitoring of CVP was performed in the clinical assessment group

Funding

National Heart, Lung, and Blood Institute