Cardiorenal Rescue Study in Acute Decompensated Heart Failure - CARRESS-HF


Cardiorenal syndrome is a common complication in patients admitted with acute decompensated heart failure (ADHF). Slow continuous ultrafiltration (SCUF) is sometimes necessary in patients with ADHF and pulmonary edema who are unresponsive to intravenous (IV) diuretics and develop cardiorenal syndrome. The current trial sought to investigate if SCUF would be superior to stepped pharmacological treatment (conventional management) in improving renal function and promoting weight loss/diuresis in patients admitted with ADHF with evidence of worsening renal function.


SCUF would be superior to conventional management with stepped pharmacological treatment in patients admitted with ADHF with evidence of worsening renal function/cardiorenal syndrome.

Study Design

  • Randomized
  • Parallel
  • Stratified

Patient Populations:

  • Age 18 or older
  • Admitted to hospital with ADHF
  • Worsened renal function with increase in Cr ≥0.3 mg/dl
  • Persistent congestion

    Number of enrollees: 15,871
    Duration of follow-up: 31 months
    Mean patient age: 60 years
    Percentage female: 19%
    New York Heart Association class: I/II (15%)


  • Cr >3.5 mg/dl
  • Alternate explanation for worsening renal function
  • SBP <90 mm Hg
  • Hematocrit >45%
  • Need for IV vasoactive drugs

Primary Endpoints:

  • Change in serum Cr and change in weight between randomization and 96 hours,
    considered as a bivariate response

Secondary Endpoints:

  • Rate of clinical decongestion
  • Measures of global well-being and dyspnea

Drug/Procedures Used:

Patients were randomized in a 1:1 fashion to either SCUF or stepped pharmacological management. Venovenous SCUF was performed with IV access at a rate of 200 ml/hr (Aquadex system 100, CHF solutions). All diuretics were stopped for the duration of SCUF, and the use of inotropes or vasodilators was prohibited. Heparin was administered in all patients to achieve a partial thromboplastin time 2-2.5 x normal values. In the stepped pharmacological management arm, IV diuretics were adjusted to achieve a urine output of 3-5 L/day.

After 48 hours, either dopamine/dobutamine (if ejection fraction [EF] <40% and systolic blood pressure [SBP] <110 mm Hg) or nitroglycerin/nesiritide (SBP >120 mm Hg or severe symptoms) were added. If urine output remained inadequate after 72 hours, these patients could undergo SCUF, hemodialysis, or receive tailored therapy based on hemodynamic measurements.

Concomitant Medications:

Before admission: Angiotensin-converting enzyme inhibitors/angiotensin-receptor blockers (54%), beta-blockers (79%), aldosterone antagonists (20%), diuretic (94%)

Principal Findings:

The trial was terminated early due to futility and an excess of adverse events in the SCUF arm. At this time, a total of 188 patients had been randomized, 94 to SCUF and 94 to pharmacological treatment. Baseline characteristics were fairly similar between the two arms. Median baseline EF was 30-35%, and approximately 60% had an ischemic cardiomyopathy. More than three quarters of these patients were hospitalized with congestive heart failure (CHF) within the past year. The median baseline creatinine (Cr) was 2.0 mg/dl, and the median qualifying increase in Cr (for definition of cardiorenal syndrome) was 0.45 mg/dl. The median duration of SCUF was 40 hours, with a crossover rate of 39% before the 96-hour assessment period was completed. The median duration of stepped pharmacological treatment was 92 hours, with a crossover rate of 6%.

The primary endpoint of change in weight and change in serum Cr 96 hours after enrollment was significantly higher in the SCUF arm as compared with the stepped pharmacological management arm (p = 0.003). This difference was driven mostly by a significant increase in serum Cr from baseline in the SCUF arm (0.23 mg/dl vs. -0.04 mg/dl, p = 0.003). Mean weight loss was similar (5.5 kg vs. 5.7 kg, p = 0.58). Mean change in Cr was similar between the two arms at 24 hours, but by 48 hours was higher in the SCUF arm (p = 0.048), which got progressively worse to 96 hours (p = 0.002). There was also a significant reduction in sodium levels from baseline to 96 hours in the SCUF arm (-2.3 vs. 0 mmol/L, p < 0.0001). Rate of clinical decongestion at 96 hours was similar (10% vs. 9%, p = 0.83).

Clinical outcomes including death at 60 days (17% vs. 14%, p = 0.55) and CHF hospitalization (26% vs. 26%, p = 0.97) were similar between the two arms. Serious adverse events at 60 days were higher with SCUF (72% vs. 57%, p = 0.03), driven by increases in the incidence of CHF, kidney failure, bleeding complications, and catheter-related complications.


Cardiorenal syndrome affects as many as one fourth to one third of patients presenting with CHF, and is associated with high morbidity and mortality. The results of the CARRESS-CHF trial indicate that SCUF is not superior to stepped pharmacological treatment in the management of patients with ADHF who have cardiorenal syndrome. Serum Cr increased rather than decreased in the SCUF arm by 48 hours. SCUF also resulted in worsening hyponatremia and a higher incidence of adverse events including some related to SCUF administration. It is possible that the worsening of serum Cr noted in this trial was due to rapid removal of fluid (200 ml/hr) with SCUF, without allowing time for re-equilibration as happens with pharamacological therapy. This could have been avoided if tailored hemodynamic therapy was used early, or other measurements of cardiac filling pressures such as intra-abdominal pressure or echocardiographic parameters (such as e/e’) were included in the treatment paradigm. Similarly, it is unknown if a slower rate of SCUF would have resulted in better outcomes; the optimal rate of SCUF in these patients remains unknown.


Bart BA, Goldsmith SR, Lee KL, et al. Ultrafiltration in decompensated heart failure with cardiorenal syndrome. N Engl J Med 2012;367:2296-2304.

Presented by Dr. Bradley Bart at the American Heart Association Scientific Sessions, Los Angeles, CA, November 6, 2012.

Clinical Topics: Anticoagulation Management, Heart Failure and Cardiomyopathies, Acute Heart Failure, Chronic Heart Failure, Heart Failure and Cardiac Biomarkers

Keywords: Pulmonary Edema, Cardio-Renal Syndrome, Sodium, Weight Loss, Body Weight, Diuretics, Heparin, Diuresis, Blood Pressure, Dopamine, Creatinine, Vasodilator Agents, Renal Dialysis, Renal Insufficiency, Hyponatremia, Dobutamine, Cardiomyopathies, Partial Thromboplastin Time, Medical Futility, Ultrafiltration, Hospitalization, Nitroglycerin, Natriuretic Peptide, Brain

< Back to Listings