IJCA-17606; No of Pages 8 International Journal of Cardiology xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials☆ Joey S.W. Kwong, Cheuk-Man Yu ⁎ Institute of Vascular Medicine, Li Ka Shing Institute of Health Sciences, Division of Cardiology, Heart Education And Research Training (HEART) Centre, Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong
a r t i c l e
i n f o
Article history: Received 27 May 2013 Received in revised form 13 January 2014 Accepted 18 January 2014 Available online xxxx Keywords: Ultrafiltration Acute decompensated heart failure Systematic review Meta-analysis
a b s t r a c t Background: Current clinical guidelines recommend ultrafiltration (UF) for patients with acute decompensated heart failure (ADHF) who are unresponsive or resistant to diuretics. We systematically reviewed the latest randomized evidence on the efficacy and safety of UF in ADHF. Methods: MEDLINE, EMBASE and the Cochrane database were searched in January 2013 for eligible randomized controlled trials (RCTs) evaluating UF in patients with ADHF. A Mantel–Haenszel random-effects model was used to calculate mean differences (MDs) and odds ratios (ORs) for continuous and dichotomous data, respectively, with 95% confidence intervals (CIs). Results: Data of 12 studies (n = 659) were meta-analyzed; follow-up duration ranged from 36 h to 12 months. Compared to control, treatment of UF was associated with significant fluid removal (MD 1.28, 95% CI 0.43 to 2.12, P = 0.003) and weight loss (MD 1.23, 95% CI 0.03 to 2.44, P = 0.04), with no significant effects on allcause mortality (OR 1.08, 95% CI 0.63 to 1.86, P = 0.77) or all-cause rehospitalization (OR 0.89, 95% CI 0.39 to 2.00, P = 0.77). No significant differences were observed in the analyses of change in serum creatinine or unscheduled medical care; analysis of adverse effects was inconclusive since only one study provided usable data. Conclusions: For patients with ADHF, UF is effective in reducing fluid retention and body weight, with no significant benefits in mortality or rehospitalization. The current limited randomized evidence highlights the need for further well-conducted randomized studies of adequate power to establish the role of UF in ADHF patients for whom conventional HF treatment is unsuccessful or contraindicated. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction An estimated 5.1 million Americans over the age of 20 years have heart failure (HF) [1]. By 2030, the prevalence of HF will increase by 25% and the consequent total cost of HF will increase almost 120% to $70 billion [1]. Acute decompensated heart failure (ADHF) is a sudden deterioration of chronic stable HF characterized by sodium retention leading to volume overload, and congestive symptoms such as edema and dyspnea occur [2,3]. ADHF is an emerging public health problem with high hospitalization rate and poor prognosis [4]. In the EuroHeart Failure Survey II (EHFS II; n = 3580), ADHF was found to be the predominant classification (62.9%) of all hospitalized acute HF cases with an inhospital mortality rate of 5.8% [5]. The large, multicenter Acute Decompensated Heart Failure National Registry (ADHERE; n = 105,388) also indicated that ADHF-related hospital admissions were associated with high inhospital mortality rate (4%), with a higher rate observed among patients admitted to intensive care settings (11%) [6]. Intravenous (IV) loop diuretics are recommended first-line treatment for
significant pulmonary congestion and fluid retention in the latest clinical guidelines [7,8]. However, the effectiveness of diuretics in ADHF is hindered by several factors such as the elimination of hypotonic urine, diuretic resistance, electrolyte disturbances and impaired glomerular filtration rate [9–11]. Extracorporeal ultrafiltration (UF) is proposed as an alternative treatment strategy for patients with ADHF who are unresponsive or resistant to loop diuretics and remain in pulmonary edema (urine output b 20 mL/h) [7]. During UF, a hydrostatic pressure gradient triggers the mechanical removal of fluid across a filter membrane and isotonic plasma water is separated from blood without affecting serum electrolytes and other solutes [9–11]. A systematic review in 2011 revealed that initial evidence on the clinical effectiveness of UF was mostly derived from small, non-randomized case series [12]. However, a number of randomized controlled trials (RCTs) have since become available and we consequently conducted a systematic review and meta-analysis of randomized studies exploring the efficacy and safety of UF in patients with ADHF. 2. Methods
☆ Both authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author. Tel.: +852 2632 3594; fax: +852 2637 5643. E-mail address: [email protected] (C.-M. Yu).
2.1. Study selection We searched MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials (CENTRAL) in January 2013 to identify eligible randomized controlled trials (RCTs)
0167-5273/$ – see front matter © 2014 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijcard.2014.01.069
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
2
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
evaluating the use of UF in patients with ADHF, using the search terms: “ultrafiltration AND heart failure”. No language restrictions were applied. Reference lists of retrieved records and relevant reviews, editorials and letters were also screened. Ongoing studies were identified in ClinicalTrials.gov (www.Clinicaltrials.gov). Studies were selected independently by two authors and disagreements were resolved by discussion. 2.2. Data extraction and quality assessment Primary outcomes included: (i) fluid removal; (ii) weight loss; (iii) all-cause mortality; and (iv) all-cause rehospitalization. Secondary outcomes were change in serum creatinine, unscheduled medical care and adverse effects. Data extraction was performed independently by two authors and disagreements were resolved by discussion. We used the Cochrane Collaboration's tool for assessing risk of bias to assess quality of included trials and performed the meta-analysis according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [13,14]. 2.3. Statistical analysis We used a Mantel–Haenszel random-effects model and calculated odd ratios (ORs) and mean differences (MDs) for dichotomous and continuous data, respectively, with their 95% confidence intervals (CIs). Results were considered to be statistically significant at P b 0.05. We explored heterogeneity using the Chi-squared (Chi2) test and I2 statistic; I2 values of ≥25%, ≥50%, and ≥75% are categorized as low, moderate and high heterogeneity, respectively (P b 0.10 for statistical significance) [13]. To detect the effect of the inclusion of small studies, a sensitivity analysis using the Mantel–Haenszel fixed-effect model was performed. We also investigated if publication status of the trial reports would affect the overall findings by excluding abstract-only data in a sensitivity analysis. A sensitivity analysis was also conducted to explore if varied duration of follow-up period would affect the overall findings. Statistical analysis was performed using the Review Manager 5.2 software (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark).
3. Results 3.1. Description of studies Our study selection process is illustrated in Fig. 1. Of the 989 records identified from literature search, 930 clearly did not meet the inclusion criteria and were thus excluded. Full-texts of the remaining 59 records were retrieved for further evaluation. A total of 13 RCTs were eligible for inclusion in our systematic review; [15–33] one study (n = 10), available as a conference abstract, met our inclusion criteria but did
not provide the number of participants per group and we thus excluded it from data analysis and categorized it as a study awaiting further classification [33]. Our final meta-analysis included data from 12 RCTs enrolling 659 participants, of whom 329 (49.9%) were randomized to UF and 330 (50.1%) to control (Table 1). Sample size of the included studies varied, ranging from 12 to 200. Included participants were mostly male (up to 88.9%) and age ranged from 49 to 78.2 years. Rate of UF ranged from 100 mL/h to 600 mL/h with a variety of UF systems employed among the included studies. Control interventions consisted of IV diuretics (7 studies, n = 261), no UF (3 studies, n = 69), and unknown standard care (2 studies: conventional management (1 study, n = 7) and standard CHF therapy (1 study, n = 20)). Follow-up duration varied among studies, ranging from 36 h to 12 months. Two studies (Chung et al. [20,21], CUORE) [22–25] were available only as conference abstracts. The overall quality of the included studies was low due to insufficient information presented by the trial authors (Fig. 2). Description of methods on randomization was only provided in three studies (CARRESS-HF [18,29], Hanna et al. [26], UNLOAD) [31,32], and only CARRESS-HF and UNLOAD described methods of allocation concealment [18,19,31,32]. However, since our outcomes of interest were objective measures we decided that the overall low quality of the included studies would not affect the findings of our meta-analysis.
3.2. Primary outcomes Data on fluid removal were available from five studies randomizing 473 participants, of whom 237 (50.1%) received UF and 236 (49.9%) received controls. Treatment of UF led to a significantly greater fluid loss as compared to control (MD 1.28, 95% CI 0.43 to 2.12, P = 0.032), with nonsignificant statistical heterogeneity across studies (Chi2: P = 0.13; I2 = 43%) (Fig. 3A). Meta-analysis of weight loss included data from six studies of 480 participants; 240 (50.0%) participants were randomized to UF and the remaining 240 randomized to controls. UF was associated with a significant net weight loss as compared to control (MD 1.23, 95% CI 0.03 to 2.44, P = 0.04) with significant, moderate
Fig. 1. Study selection flow diagram.
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
Study [year]
N
UF system
Concomitant HF therapy
Control
Mean age (yr) [% male]
Cr (mg/dL)
LVEF (%)
Follow-up
Agostoni et al. [1993] [15]
36
Gambro System AK 10
Yes
No UF
?
16
Gambro System AK 10
Yes
Agostoni et al. [1995] [17]
42
Gambro System AK 10
Yes
IV furosemide (bolus 160 mg, then 1 mg/min infusion) No UF
UF: 23.8 ± 2.1 Control: 24.1 ± 1.8 ?
180 days
Agostoni et al. [1994] [16]
UF: range 49–69 [88.9] Control: range 50–74 [77.8] UF: 58 ± 1.6 [87.5] Control: 62 ± 2 [87.5]
?
188
Aquadex System 100
No
IV diuretics
Chung et al. [2010] [20,21]
16
?
No
IV furosemide infusion
CUORE [2011] [22–25]
12
Dedyca
?
Conventional management
Hanna et al. [2012] [26]
36
NxStage System One
IV diuretics
Pepi et al. [1993] [27]
24
Gambro System AK10
Stopped except spironolactone (≤25 mg/day) Yes
UF: 27.2 ± 2.8/26.8 ± 2.9b Control: 24.2 ± 1.5/24.6 ± 1.7b UF: 30 (20–52) Control: 35 (25–55) UF: 22 ± 8 Control: 26 ± 11 UF: 39 ± 1 Control: 35 ± 5 UF: 19 ± 9 Control: 18 ± 6
3 months
CARRESS-HF [2012] [18,19]
UF: 58 ± 3/63 ± 3b Control: 60 ± 4/61 ± 4b UF: 69 (61–78) [78] Control: 66 (57–78) [72] UF: 69 ± 14 Control: 74 ± 12 UF: 78.2 ± 2 Control: 72.8 ± 7 UF: 60 ± 9.1 [84.2] Control: 59 ± 15.5 [76]
RAPID-CHF [2005] [28]
40
No
Standard CHF therapy
Rogers et al. [2008] [29]
19
Simple Access Fluid Extraction (SAFE) UFC 100 Console Aquadex System 100
Yes
IV furosemide
ULTRADISCO [2011] [30]
30
PRISMA System
Yes
UNLOAD[2007] [31,32]
200
Aquadex System 100
Yes
IV furosemide infusion (initial dose 250 mg/24 h) IV diuretics
a b c d
No UF
?
UF: 1.90 (1.57–2.37) Control: 2.09 (1.71–2.65) UF: 1.9 ± 0.6 Control: 1.4 ± 0.5 ? UF: 1.6 ± 0.7 Control: 1.7 ± 0.8
UF: 57 ± 5 Control: 56 ± 5 UF: 67.5 [70] Control: 69.5 [70]
?
UF: 64 ± 15 [78] Control: 54 ± 16 [60] UF: 72.4 ± 14.1 [87] Control: 65.8 ± 18.4 [87] UF: 62 ± 15 Control: 63 ± 14
UF: 1.8 ± 0.8 Control: 1.6 ± 0.8 UF: 2.22 ± 0.75 Control: 1.86 ± 0.63 UF: 1.5 ± 0.5 Control: 1.5 ± 0.5
UF: 1.6 Control: 1.8
3 months
60 days 90 days 12 months 90 days
UF: 23.8 ± 8.7 Control: 24.2 ± 9 UF: 69%c Control: 78%c
90 days
?
48 h
UF: 34 ± 19.9 Control: 30 ± 13.4 UF: 71%d Control: 70%d
36 h
30 days
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
Table 1 Characteristics of included studies.a
90 days
Values are mean ± standard deviation or median (interquartile range) unless indicated otherwise. Participants were further stratified into two groups according to whether there was a N10% increase in peak VO2 at 3 months. Percentage of participants with LVEF b 40%. Percentage of participants with LVEF ≤ 40%.
3
4
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
analysis of adverse events was inconclusive since usable data were only available from one study (CARRESS-HF) [18,19]. 3.4. Sensitivity analysis Findings of our sensitivity analyses using the fixed-effect model to test for potential small-study effects were consistent with those of the overall random-effect analyses (Table 2). Sensitivity analyses exploring the potential influence of publication status by excluding data from Chung et al. and CUORE also showed similar results as those of the overall analyses (Table 2); [20–25] the impact of the varied follow-up durations on all-cause mortality was also explored by excluding data from acute studies (Rogers et al. and ULTRADISCO) [29,30], and findings of this sensitivity analysis were found to be comparable to the overall analysis (Table 2). 4. Discussion
Fig. 2. Assessment of risk of bias of included studies. Low risk of bias (green circles), unclear risk of bias (yellow circles) and high risk of bias (red circles). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
statistical heterogeneity among the five studies (Chi [2]: P = 0.03; I [2] = 59%) (Fig. 3B). All-cause deaths occurred in 32 of 329 (9.7%) participants allocated to UF and in 29 of 330 (8.8%) participants allocated to control. There was no significant difference between groups (OR 1.08, 95% CI 0.63 to 1.86, P = 0.77) (Fig. 4A). All-cause rehospitalization was reported in 71 of 203 (35.0%) participants treated with UF and in 74 of 204 (36.3%) participants treated with control. No significant difference was observed between the UF and control groups (OR 0.89, 95% CI 0.39 to 2.00, P = 0.77) (Fig. 4B). 3.3. Secondary outcomes There were no significant differences in the analyses of change in serum creatinine (MD 0.01, 95% CI -0.30 to 0.32; P = 0.94) or unscheduled medical care (OR 1.07, 95% CI 0.31 to 3.69; P = 0.92). Meta-
This meta-analysis of 12 studies shows that UF is effective in reducing volume overload and weight loss, with no benefits in all-cause mortality or rehospitalization. The nonsignificant differences in renal function and incidence of unscheduled medical care are inconclusive due to limited trial data. Safety profile of UF remains to be ascertained due to a lack of adverse event data in the included studies. Currently available RCTs exploring the effects of UF in ADHF are of small size and overall methodological quality was assessed to be low. Although statistical heterogeneity was observed only in the analysis of weight loss (Chi2: P = 0.03; I2 = 59%), we acknowledge the substantial clinical heterogeneity among our included studies, ranging from follow-up duration to outcomes of interest and how the outcomes were defined. Although follow-up duration was found to have minimal impact on our analysis of all-cause mortality, it is important to highlight that our analyses of fluid loss, weight loss and change in creatinine pooled study data that were obtained at varied time points (Table 3). The cut-off point of when these clinically important outcomes are measured is crucial in accurate determination of the efficacy of UF in improving clinical congestion and safety of UF in respect to renal function. We therefore call for the need of consistent cut-off time points of when outcomes are assessed in future studies. Moreover, different measures of renal function were used in the 12 included studies, which greatly contributed to the inconclusive findings of our analyses. In fact, use of serum creatinine levels or changes in serum creatinine (e.g. CARRESS-HF) as outcome measures of renal function among our included studies may not accurately reflect renal status [34]. In clinical practice, serum creatinine clearance and GFR are used as tools for renal function assessment. However, only two of our included studies (CARRESS-HF, Rogers et al.) [18,19,29] described mean changes in GFR and none provided information on baseline or follow-up serum creatinine clearance. The use of accurate and consistent measures for assessing renal function in future clinical trials studying the role of UF in ADHF is essential as it would allow for future analyses to effectively test for the effects of UF on renal function. The significant weight loss and fluid removal as well as lack of effects on serum creatinine associated with the treatment of UF as observed in our analyses are contrary to those reported by the most recent randomized study in the field, CARRESS-HF [18,19]. This randomized trial of 188 patients with ADHF and worsened renal function (defined as an elevated serum creatinine level of N 0.3 mg/dL within 12 weeks before or 10 days after admission for HF) showed that UF as compared to stepped pharmacologic therapy did not have a significant impact on fluid loss or weight loss at 96 h but led to an increase in serum creatinine; however, differences between groups with respect to mortality and rehospitalization were nonsignificant. Findings of our meta-analysis and those of CARRESS-HF illustrate that, despite significant relief of congestion with or without preservation of renal function, UF did not lead to an evident reduction in mortality or rehospitalization rates. The relationship between reduction of congestion, variations in
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
5
Fig. 3. Forest plot of (A) fluid removal; and (B) weight loss.
renal function and clinical outcomes remains unclear. In the Diuretic Optimization Strategies Evaluation (DOSE) trial [35], a significantly higher number of patients with ADHF receiving high-dose diuretic strategy (total daily IV furosemide dose 2.5 times their totally daily oral loop diuretic dose) experienced an significant increase in serum creatinine of N0.3 mg/dL as compared to those receiving low-dose diuretic strategy (23% vs. 14%; P = 0.04), greater weight loss (− 8.7 ± 8.5 lb vs. − 6.8 ± 7.8 lb; P = 0.01) and fluid loss (4899 ± 3479 mL vs. 3575 ± 2635 mL; P = 0.001), and yet the composite end point of mortality, rehospitalization and emergency department visit was comparable between groups (HR 0.83, 95% CI 0.60–1.16; P = 0.28). A single-center observational study by Patarroyo and colleagues showed that [36], in 63 patients with ADHF and worsening renal function and who were
resistant to standard medical therapy, UF significantly reduced body weight at 48-h follow-up (102 ± 25 kg vs. 99 ± 23 kg, P b 0.0001) but there was no change in serum creatinine levels (2.2 ± 0.9 mg/dL vs. 2.4 ± 1 mg/dL, P = 0.12), with somewhat alarming inhospital mortality rate of 30% and overall one-year mortality rate of 70%. Successful treatment of congestion thus cannot simply be translated into renal or clinical improvement, and ongoing research is necessary to determine a risk stratification strategy to carefully identify likely responders who might benefit from UF so to achieve a reasonable balance of congestion relief and clinical improvement. The concept of mechanical removal of excess fluid by UF was first proposed nearly 40 years ago by Silverstein and colleagues in five volume-overloaded patients on chronic hemodialysis [37]. Subsequent
Fig. 4. Forest plot of (A) all-cause mortality; and (B) all-cause rehospitalization.
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
6
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
Table 2 Results of sensitivity analyses of (A) primary outcomes; and (B) secondary outcomes. (A) Net fluid removal (mL) MD (95% CI)
Net weight loss (kg) MD (95% CI)
All-cause mortality OR (95% CI)
All-cause rehospitalization OR (95% CI)
Analytical method Random-effects Fixed-effect
1.28 (0.43, 2.12)⁎ 1.24 (0.69, 1.78)⁎
1.23 (0.03, 2.44)⁎ 1.41 (0.76, 2.06)⁎
1.08 (0.63, 1.86) 1.11 (0.65, 1.88)
0.89 (0.39, 2.00) 0.95 (0.63, 1.44)
Publication status All trials Chung et al. and CUORE excluded
1.28 (0.43, 2.12)⁎ NA
1.23 (0.03, 2.44)⁎ 1.52 (0.11, 2.93)⁎
1.08 (0.63, 1.86) 1.11 (0.64, 1.92)
0.89 (0.39, 2.00) 0.96 (0.40, 2.35)
Follow-up duration Rogers et al. and ULTRADISCO excluded
NA
NA
1.08 (0.63, 1.86)
NA
Analysis
(B) Analysis
Change in serum creatinine (mg/dL) MD (95% CI)
Unscheduled medical care OR (95% CI)
Analytical method Random-effects Fixed-effect
0.01 (−0.30, 0.32) 0.09 (−0.05, 0.22)
1.07 (0.31, 3.69) 0.83 (0.51, 1.34)
Publication status All trials Chung et al. and CUORE excluded
0.01 (−0.30, 0.32) 0.08 (−0.30, 0.46)
1.07 (0.31, 3.69) NA
Abbreviations: CI, confidence interval; MD, mean difference; NA, not applicable; OR, odds ratio * P b 0.05
small observational studies in the 1980s showed early promises of UF in fluid and weight loss, symptomatic improvement with increased response to diuretics in patients with refractory congestive HF [38,39]. However, as illustrated by our literature search, randomized studies of moderate size such as the UNLOAD study were only recently published. During this research gap, much has advanced in the development of different UF systems and physicians as well as patients are faced with an increasing number of options. For example, the FDA-approved Aquadex FlexFlow™ system (Gambro AB, Lund, Sweden) and the Dedyca® system (Bellco, Mirandola, Italy) both utilize a simplistic design and do not require expert management by nephrologists [40]. The Aquadex FlexFlow™ is a portable UF device performed via peripheral venous access whereas the Dedyca® system uses a femoral venous approach. Although a direct head-to-head comparative study to explore the relative effects of the different UF systems is not yet available, data have shown different outcomes associated the use of UF via peripheral vs. femoral access. Liang et al. used the Aquadex FlexFlow™ system in 11 patients with refractory HF who are resistant to diuretics or at risk of resistance Table 3 Cut-off time points of measurement of fluid loss, weight loss and change in creatinine clearance. Study
Time point
Fluid loss CARRESS-HF 2012 Hanna 2012 Rogers 2008 ULTRADISCO 2011 UNLOAD 2007
From randomization to 96 h During intervention period From baseline to 48 h From baseline to 36 h From randomization to 48 h
Weight loss CARRESS-HF 2012 Chung 2010 CUORE (short term) Hanna 2012 Rogers 2008 UNLOAD 2007
From randomization to 96 h ? At discharge During intervention period From baseline to 48 h From randomization to 48 h
Change in serum creatinine CARRESS-HF 2012 CUORE (short term) Rogers 2008
From randomization to 96 h At discharge From baseline to 48 h
with an aim to remove 4 L of fluid per a 8-h UF session, and reported a six-month mortality rate of 55%, 45% of patients experienced a worsened renal function (increase of serum creatinine of N0.3 mg/mL) and 45% of patients required dialysis for persistent diuretic-resistant volume overload and uremic symptoms [41]. In contrast, a study by De Maria and colleagues using the Dedyca® system among 42 patients with congestive HF with LVEF b 40% and diuretic resistance gave a more favorable six-month mortality rate of 26%, only 14% of patients were found with worsened renal function (serum creatinine N3 mg/dL and/or glomerular filtration rate (GFR) b30 mL/min) and only 1 patient (2.4%) required dialysis for persistent diuretic-resistant volume overload and uremic symptoms [42]. The similar baseline characteristics of the enrolled participants of the studies by Liang et al. (age: 70 years; LVEF: 43%; serum creatinine: 2.2 mg/dL; GFR: 38 mL/min; pre-UF furosemide dose: 258 mg/day) and by De Maria et al. (age: 69 ± 13 years; LVEF: 32 ± 5%; serum creatinine: 2 ± 0.6 mg/dL; GFR: 58 ± 27 mL/min; pre-UF furosemide dose: 250 mg/day) and the different clinical outcomes reported suggest that there may be potential device-specific effects in using UF for patients with ADHF. Direct comparison of the effects of UF between the different systems should be evaluated using a wellplanned head-to-head study. However, findings of the aforementioned observational studies illustrate potential influences of treatment regimens (type of system, filtration rate, duration) on the efficacy and safety of UF. It is worth highlighting findings of an interesting study by Costanzo et al., which explored if initiating UF in the early phase of hospitalization would result in euvolemia and early discharge (defined as discharge in ≤3 days) [43]. In a group of 20 patients with ADHF, fluid overload and diuretic resistance/renal insufficiency, UF was initiated within 4.7 ± 3.5 h of hospitalization prior to the administration of IV diuretics. Early UF effectively reduced length of hospital stay (average 3.7 ± 1.8 days), for which 60% of patients were discharged in ≤3 days; quality of life (Minnesota Living With Heart Failure Questionnaire (MLWHFQ)) and global assessment improved from pre-treatment stage (MLWHFQ score: 70 ± 18; global assessment: 5.7 ± 1.3) to 90day follow-up (MLWHFQ score: 51 ± 27; global assessment: 2.5 ± 1.5). Besides the timing of UF initiation, filtration rate has also been raised as an important factor, since aggressive volume removal exceeding the plasma refill rate may lead to intravascular volume depletion,
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
hypotension and renal hypoperfusion [44,45]. It would be of interest and relevance to conduct a subgroup analysis investigating the outcomes associated with different UF strategies. However, information on the UF treatment regimens employed by our included studies is limited. Detailed description of UF regimens is essential to determine the most effective and safest access site (peripheral vs. femoral), treatment duration, volume removal rate and quantity. Furthermore, randomized data on potential adverse events associated with UF are suboptimal, with just two of our included studies (CARRESS-HF [18,19], UNLOAD) [31,32] reporting adverse events and only CARRESS-HF provided event numbers per arm; [18,19] CARRESS-HF showed that the incidence of serious adverse events was significantly higher in the UF arm compared to the pharmacological arm (72% vs. 57%, P = 0.03); this high percentage was mainly attributed to heart and renal failure (33% and 18%, respectively) [18,19]. We thus call for the need of consistent and transparent reporting of complications associated with the treatment of UF in future studies to allow for comprehensive analysis of adverse effects. Three ongoing RCTs comparing diuretics and UF in patients with ADHF were identified in our literature search, all of which are estimated to complete in late 2013. The Aquapheresis Versus Intravenous Diuretics and Hospitalizations for Heart Failure (AVOID-HF) study aims to enroll 810 participants with a primary outcome measure of time to first HF event at 90 days; [46] the REverse WOrsening Renal Function in Decompensated Heart Failure (REWORD-HF) estimates to recruit 183 participants with a primary outcome of changes in dyspnea, weight loss, GFR and brain natriuretic peptide (BNP); [47] and the third study aims to assess myocardial blood flow in an estimated study population of 30 [48]. Results of these ongoing studies are much anticipated and future updated meta-analyses incorporating their data will provide further insights to the precise efficacy and safety profiles of UF in patients with ADHF. In addition, the suboptimal methodological quality demonstrated by the currently available RCTs not only highlights the issue of incomplete reporting of trial methodology but also emphasizes the importance of adequate description of trial design and conduct, especially on the domains of sequence generation and allocation concealment. Conclusions of any meta-analyses rely on the validity and integrity of the included studies. Future meta-analyses should therefore pay particular attention to the assessment of risk of bias in included studies in order to provide better insights to the extent to which the conclusions can be drawn. 4.1. Study limitations Limitations of our meta-analysis include the marked clinical heterogeneity and moderate statistical heterogeneity among the included studies in terms of UF system used, sample size, follow-up duration and reporting of outcomes. The diverse randomized evidence available indicates that findings of our analyses should be interpreted with caution. 5. Conclusions Currently available knowledge on the efficacy and safety of UF in patients with ADHF is inconclusive, where published studies are of small size and reporting of relevant outcome measures is suboptimal. Wellconducted randomized trials with adequate power and carefully selected endpoints are much warranted to enrich the existing body of evidence on the role of UF in patients with refractory HF and resistance to standard diuretic therapy or in whom such therapy is contraindicated. Acknowledgment This study was supported by a research grant from the University Grants Committee of Hong Kong (RGC Collaborative Research Fund 2010/11: CUHK9/CRF/10). We also acknowledge support from the Lui Che Woo Foundation.
7
References [1] Go AS, Mozaffarian D, Roger VL, et al; on behalf of the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics — 2013 update: a report from the American Heart Association. Circulation 2013;127:e6-245. [2] Lindenfeld J, Albert NM, Boehmer JP, et al. HFSA 2010 comprehensive heart failure practice guideline. J Card Fail 2010;16:e1-194. [3] Onwuanyi A, Taylor M. Acute decompensated heart failure: pathophysiology and treatment. Am J Cardiol 2007;99(6B):25D–30D. [4] Felker GM, Mentz RJ. Diuretics and ultrafiltration in acute decompensated heart failure. J Am Coll Cardiol 2012;59:2145–53. [5] Nieminen MS, Brutsaert D, Dickstein K, et al. EuroHeart Survey Investigators. EuroHeart Failure Survey II (EHFS II): a survey on hospitalized acute heart failure patients: description of population. Eur Heart J 2006;27:2725–36. [6] Adams Jr KF, Fonarow GC, Emerman CL, et al. ADHERE scientific advisory committee and investigators. Characteristics and outcomes of patients hospitalized for heart failure in the United States: rationale, design, and preliminary observations from the first 100,000 cases in the Acute Decompensated Heart Failure National Registry (ADHERE). Am Heart J 2005;149:209–16. [7] McMurray JJ, Adamopoulos S, Anker SD, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2012. Eur J Heart Fail 2012;14:803–69. [8] Jessup M, Abraham WT, Casey DE, et al; writing on behalf of the 2005 Guideline Update for the Diagnosis and Management of Chronic Heart Failure in the Adult Writing Committee. 2009 Focused update: ACCF/AHA guidelines for the diagnosis and management of heart failure in adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation 2009;119:1977–2016. [9] Bart BA. Treatment of congestion in congestive heart failure: ultrafiltration is the only rational initial treatment of volume overload in decompensated heart failure. Circ Heart Fail 2009;2:499–504. [10] Wertman BM, Gura V, Schwarz ER. Ultrafiltration for the management of acute decompensated heart failure. J Card Fail 2008;14:754–9. [11] Costanzo MR, Jessup M. Treatment of congestion in heart failure with diuretics and extracorporeal therapies: effects on symptoms, renal function, and prognosis. Heart Fail Rev 2012;17:313–24. [12] Fiaccadori E, Regolisti G, Maggiore U, et al. Ultrafiltration in heart failure. Am Heart J 2011;161:439–49. [13] Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration; 2011 http://handbook.cochrane.org/. [Accessed 25 May 2013]. [14] Moher D, Liberati A, Tetzlaff J, Altman DG; the PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097. [15] Agostoni PG, Marenzi GC, Pepi M, et al. Isolated ultrafiltration in moderate congestive heart failure. J Am Coll Cardiol 1993;21:424–31. [16] Agostoni P, Marenzi G, Lauri G, et al. Sustained improvement in functional capacity after removal of body fluid with isolated ultrafiltration in chronic cardiac insufficiency: failure of furosemide to provide the same result. Am J Med 1994;96:191–9. [17] Agostoni PG, Marenzi GC, Sganzerla P, et al. Lung–heart interaction as a substrate for the improvement in exercise capacity after body fluid volume depletion in moderate congestive heart failure. Am J Cardiol 1995;76:793–8. [18] Bart BA, Goldsmith SR, Lee KL, et al; for the Heart Failure Clinical Research Network. Ultrafiltration in decompensated heart failure with cardiorenal syndrome. N Engl J Med 2012;367:2296–304. [19] Bart BA, Goldsmith SR, Lee KL, et al. Cardiorenal rescue study in acute decompensated heart failure: rationale and design of CARRESS-HF, for the Heart Failure Clinical Research Network. J Card Fail 2012;18:176–82. [20] Chung E, Bartone C, Menon S, Mazur W, O'Brien T, Kereiakes D. Target weight guided treatment of acute heart failure using ultrafiltration or usual care: results of a randomized pilot study. Presented at the European Society of Cardiology (ESC) congress; 2010 [28 Aug–01 Sep 2010, Stockholm, Sweden]. [21] Bartone C, Menon SG, Kereiakes DJ, et al. Target weight guided treatment of acute heart failure using ultrafiltration or usual care: results of a randomized pilot study. J Card Fail 2010;16(8S):S106. [22] Cosentino ER, Rinaldi ER, Degli Esposti D, et al. Preliminary report on the effects of ultrafiltration in severe HF refractory to conventional diuretic therapy: the Continuous Ultrafiltration for cOngestive heaRt failurE (CUORE) trial. Eur J Heart Fail Suppl 2011;10:S112. [23] Ferramosca E, Mancini E, Cosentino E, et al. Preliminary report on the effects of ultrafiltration in severe heart failure: the Continuous Ultrafiltration for cOngestive heaRt failurE (CUORE) trial. NDT Plus 2009;2(Suppl. 2):ii427. [24] Muratori M, Agostoni PG, Marenzi G, et al. Preliminary report on the effects of ultrafiltration in severe heart failure: the Continuous Ultrafiltration for cOngestive heaRt failurE (CUORE) trial. Eur J Heart Fail Suppl 2008;7(Suppl. 1):72. [25] Ferramosca E, Cosentino ER, Muratori M, et al; on behalf of the CUORE Trial Investigators. Continuous ultrafiltration for congestive heart failure (CUORE) trial: Risultati preliminari sugli effetti dell'ultrafiltrazione isolata nel trattamento dello scompenso cardiaco severo. G Ital Nefrol 2008;25(S-43):S31. [26] Hanna MA, Tang WH, Teo BW, et al. Extracorporeal ultrafiltration vs. conventional diuretic therapy in advanced decompensated heart failure. Congest Heart Fail 2012;18:54–63. [27] Pepi M, Marenzi GC, Agostoni PG, et al. Sustained cardiac diastolic changes elicited by ultrafiltration in patients with moderate congestive heart failure: pathophysiological correlates. Br Heart J 1993;70:135–40.
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
8
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
[28] Bart BA, Boyle A, Bank AJ, et al. Ultrafiltration versus usual care for hospitalized patients with heart failure: the relief for acutely fluid-overloaded patients with decompensated congestive heart failure (RAPID-CHF) trial. J Am Coll Cardiol 2005;46:2043–6. [29] Rogers HL, Marshall J, Bock J, et al. A randomized, controlled trial of the renal effects of ultrafiltration as compared to furosemide in patients with acute decompensated heart failure. J Card Fail 2008;14:1–5. [30] Giglioli C, Landi D, Cecchi E, et al. Effects of ULTRAfiltration vs. DIureticS on clinical, biohumoral and haemodynamic variables in patients with deCOmpensated heart failure: the ULTRADISCO study. Eur J Heart Fail 2011;13:337–46. [31] Costanzo MR, Guglin ME, Saltzberg MT, et al; for the UNLOAD Trial Investigators. Ultrafiltration versus intravenous diuretics for patients hospitalized for acute decompensated heart failure. J Am Coll Cardiol 2007;49:675–83. [32] Cleland JG, Coletta AP, Nikitin NP, Clark AL. Clinical trials update from the American College of Cardiology: Darbepoetin alfa, ASTEROID, UNIVERSE, paediatric carvedilol, UNLOAD and ICELAND. Eur J Heart Fail 2006;8:326–9. [33] Stein AC, Mostarda C, Alves B, et al. Ultrafiltration treatment in decompensated heart failure: changes in heart rate variability and survival. Hypertension 2011;58:e64. [34] Kazory A, Ross EA. Ultrafiltration for decompensated heart failure: renal implications. Heart 2009;95:1047–51. [35] Felker GM, Lee KL, Bull DA, et al. NHLBI heart failure clinical research network. Diuretic strategies in patients with acute decompensated heart failure. N Engl J Med 2011;364:797–805. [36] Patarroyo M, Wehbe E, Hanna M, et al. Cardiorenal outcomes after slow continuous ultrafiltration therapy in refractory patients with advanced decompensated heart failure. J Am Coll Cardiol 2012;60:1906–12. [37] Silverstein ME, Ford CA, Lysaght MJ, Henderson LW. Treatment of severe fluid overload by ultrafiltration. N Engl J Med 1974;291:747–51. [38] Simpson IA, Rae AP, Simpson K, et al. Ultrafiltration in the management of refractory congestive heart failure. Br Heart J 1986;55:344–7.
[39] DiLeo M, Pacitti A, Bergerone S, et al. Ultrafiltration in the treatment of refractory congestive heart failure. Clin Cardiol 1988;11:449–52. [40] Wańkowicz Z, Próchnicka A, Olszowska A, Baczyński D, Krzesiński P, Dziuk M. Extracorporeal versus peritoneal ultrafiltration in diuretic-resistant congestive heart failure—a review. Med Sci Monit 2011;17:RA271–81. [41] Liang KV, Hiniker AR, Williams AW, Karon BL, Greene EL, Redfield MM. Use of a novel ultrafiltration device as a treatment strategy for diuretic resistant, refractory heart failure: initial clinical experience in a single center. J Card Fail 2006;12:707–14. [42] De Maria E, Pignatti F, Patrizi G, Benenati PM, Ricci S, Cappelli S. Ultrafiltration for the treatment of diuretic-resistant, recurrent, acute decompensated heart failure: experience in a single center. J Cardiovasc Med 2010;11:599–604. [43] Costanzo MR, Saltzberg M, O'Sullivan J, Sobotka P. Early ultrafiltration in patients with decompensated heart failure and diuretic resistance. J Am Coll Cardiol 2005;46:2047–51. [44] Schroeder KL, Sallustio JE, Ross EA. Continuous haematocrit monitoring during intradialytic hypotension: precipitous decline in plasma refill rates. Nephrol Dial Transplant 2004;19:652–6. [45] Kazory A, Ross EA. Contemporary trends in the pharmacological and extracorporeal management of heart failure: a nephrologic perspective. Circulation 2008;117:975–83. [46] NCT01474200. Study of Heart failure hospitalizations after aquapheresis therapy compared to intravenous (IV) diuretic treatment (AVOID-HF). http://www. clinicaltrials.gov/ct/show/NCT01474200 . [accessed 25 May 2013]. [47] NCT01140399. REWORD-HF REverse WOrsening Renal Function in Decompensated Heart Failure. http://www.clinicaltrials.gov/ct2/show/NCT01140399 . [accessed 25 May 2013]. [48] NCT01457053. Assessment of Coronary Flow Reserve in Heart Failure Patients After Ultrafiltration Versus Diuretics. http://www.clinicaltrials.gov/ct2/show/NCT01457053 . [accessed 25 May 2013].
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
Contents lists available at ScienceDirect
International Journal of Cardiology journal homepage: www.elsevier.com/locate/ijcard
Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials☆ Joey S.W. Kwong, Cheuk-Man Yu ⁎ Institute of Vascular Medicine, Li Ka Shing Institute of Health Sciences, Division of Cardiology, Heart Education And Research Training (HEART) Centre, Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, N.T., Hong Kong
a r t i c l e
i n f o
Article history: Received 27 May 2013 Received in revised form 13 January 2014 Accepted 18 January 2014 Available online xxxx Keywords: Ultrafiltration Acute decompensated heart failure Systematic review Meta-analysis
a b s t r a c t Background: Current clinical guidelines recommend ultrafiltration (UF) for patients with acute decompensated heart failure (ADHF) who are unresponsive or resistant to diuretics. We systematically reviewed the latest randomized evidence on the efficacy and safety of UF in ADHF. Methods: MEDLINE, EMBASE and the Cochrane database were searched in January 2013 for eligible randomized controlled trials (RCTs) evaluating UF in patients with ADHF. A Mantel–Haenszel random-effects model was used to calculate mean differences (MDs) and odds ratios (ORs) for continuous and dichotomous data, respectively, with 95% confidence intervals (CIs). Results: Data of 12 studies (n = 659) were meta-analyzed; follow-up duration ranged from 36 h to 12 months. Compared to control, treatment of UF was associated with significant fluid removal (MD 1.28, 95% CI 0.43 to 2.12, P = 0.003) and weight loss (MD 1.23, 95% CI 0.03 to 2.44, P = 0.04), with no significant effects on allcause mortality (OR 1.08, 95% CI 0.63 to 1.86, P = 0.77) or all-cause rehospitalization (OR 0.89, 95% CI 0.39 to 2.00, P = 0.77). No significant differences were observed in the analyses of change in serum creatinine or unscheduled medical care; analysis of adverse effects was inconclusive since only one study provided usable data. Conclusions: For patients with ADHF, UF is effective in reducing fluid retention and body weight, with no significant benefits in mortality or rehospitalization. The current limited randomized evidence highlights the need for further well-conducted randomized studies of adequate power to establish the role of UF in ADHF patients for whom conventional HF treatment is unsuccessful or contraindicated. © 2014 Elsevier Ireland Ltd. All rights reserved.
1. Introduction An estimated 5.1 million Americans over the age of 20 years have heart failure (HF) [1]. By 2030, the prevalence of HF will increase by 25% and the consequent total cost of HF will increase almost 120% to $70 billion [1]. Acute decompensated heart failure (ADHF) is a sudden deterioration of chronic stable HF characterized by sodium retention leading to volume overload, and congestive symptoms such as edema and dyspnea occur [2,3]. ADHF is an emerging public health problem with high hospitalization rate and poor prognosis [4]. In the EuroHeart Failure Survey II (EHFS II; n = 3580), ADHF was found to be the predominant classification (62.9%) of all hospitalized acute HF cases with an inhospital mortality rate of 5.8% [5]. The large, multicenter Acute Decompensated Heart Failure National Registry (ADHERE; n = 105,388) also indicated that ADHF-related hospital admissions were associated with high inhospital mortality rate (4%), with a higher rate observed among patients admitted to intensive care settings (11%) [6]. Intravenous (IV) loop diuretics are recommended first-line treatment for
significant pulmonary congestion and fluid retention in the latest clinical guidelines [7,8]. However, the effectiveness of diuretics in ADHF is hindered by several factors such as the elimination of hypotonic urine, diuretic resistance, electrolyte disturbances and impaired glomerular filtration rate [9–11]. Extracorporeal ultrafiltration (UF) is proposed as an alternative treatment strategy for patients with ADHF who are unresponsive or resistant to loop diuretics and remain in pulmonary edema (urine output b 20 mL/h) [7]. During UF, a hydrostatic pressure gradient triggers the mechanical removal of fluid across a filter membrane and isotonic plasma water is separated from blood without affecting serum electrolytes and other solutes [9–11]. A systematic review in 2011 revealed that initial evidence on the clinical effectiveness of UF was mostly derived from small, non-randomized case series [12]. However, a number of randomized controlled trials (RCTs) have since become available and we consequently conducted a systematic review and meta-analysis of randomized studies exploring the efficacy and safety of UF in patients with ADHF. 2. Methods
☆ Both authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. ⁎ Corresponding author. Tel.: +852 2632 3594; fax: +852 2637 5643. E-mail address: [email protected] (C.-M. Yu).
2.1. Study selection We searched MEDLINE, EMBASE, and the Cochrane Central Register of Controlled Trials (CENTRAL) in January 2013 to identify eligible randomized controlled trials (RCTs)
0167-5273/$ – see front matter © 2014 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.ijcard.2014.01.069
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
2
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
evaluating the use of UF in patients with ADHF, using the search terms: “ultrafiltration AND heart failure”. No language restrictions were applied. Reference lists of retrieved records and relevant reviews, editorials and letters were also screened. Ongoing studies were identified in ClinicalTrials.gov (www.Clinicaltrials.gov). Studies were selected independently by two authors and disagreements were resolved by discussion. 2.2. Data extraction and quality assessment Primary outcomes included: (i) fluid removal; (ii) weight loss; (iii) all-cause mortality; and (iv) all-cause rehospitalization. Secondary outcomes were change in serum creatinine, unscheduled medical care and adverse effects. Data extraction was performed independently by two authors and disagreements were resolved by discussion. We used the Cochrane Collaboration's tool for assessing risk of bias to assess quality of included trials and performed the meta-analysis according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement [13,14]. 2.3. Statistical analysis We used a Mantel–Haenszel random-effects model and calculated odd ratios (ORs) and mean differences (MDs) for dichotomous and continuous data, respectively, with their 95% confidence intervals (CIs). Results were considered to be statistically significant at P b 0.05. We explored heterogeneity using the Chi-squared (Chi2) test and I2 statistic; I2 values of ≥25%, ≥50%, and ≥75% are categorized as low, moderate and high heterogeneity, respectively (P b 0.10 for statistical significance) [13]. To detect the effect of the inclusion of small studies, a sensitivity analysis using the Mantel–Haenszel fixed-effect model was performed. We also investigated if publication status of the trial reports would affect the overall findings by excluding abstract-only data in a sensitivity analysis. A sensitivity analysis was also conducted to explore if varied duration of follow-up period would affect the overall findings. Statistical analysis was performed using the Review Manager 5.2 software (The Nordic Cochrane Centre, The Cochrane Collaboration, Copenhagen, Denmark).
3. Results 3.1. Description of studies Our study selection process is illustrated in Fig. 1. Of the 989 records identified from literature search, 930 clearly did not meet the inclusion criteria and were thus excluded. Full-texts of the remaining 59 records were retrieved for further evaluation. A total of 13 RCTs were eligible for inclusion in our systematic review; [15–33] one study (n = 10), available as a conference abstract, met our inclusion criteria but did
not provide the number of participants per group and we thus excluded it from data analysis and categorized it as a study awaiting further classification [33]. Our final meta-analysis included data from 12 RCTs enrolling 659 participants, of whom 329 (49.9%) were randomized to UF and 330 (50.1%) to control (Table 1). Sample size of the included studies varied, ranging from 12 to 200. Included participants were mostly male (up to 88.9%) and age ranged from 49 to 78.2 years. Rate of UF ranged from 100 mL/h to 600 mL/h with a variety of UF systems employed among the included studies. Control interventions consisted of IV diuretics (7 studies, n = 261), no UF (3 studies, n = 69), and unknown standard care (2 studies: conventional management (1 study, n = 7) and standard CHF therapy (1 study, n = 20)). Follow-up duration varied among studies, ranging from 36 h to 12 months. Two studies (Chung et al. [20,21], CUORE) [22–25] were available only as conference abstracts. The overall quality of the included studies was low due to insufficient information presented by the trial authors (Fig. 2). Description of methods on randomization was only provided in three studies (CARRESS-HF [18,29], Hanna et al. [26], UNLOAD) [31,32], and only CARRESS-HF and UNLOAD described methods of allocation concealment [18,19,31,32]. However, since our outcomes of interest were objective measures we decided that the overall low quality of the included studies would not affect the findings of our meta-analysis.
3.2. Primary outcomes Data on fluid removal were available from five studies randomizing 473 participants, of whom 237 (50.1%) received UF and 236 (49.9%) received controls. Treatment of UF led to a significantly greater fluid loss as compared to control (MD 1.28, 95% CI 0.43 to 2.12, P = 0.032), with nonsignificant statistical heterogeneity across studies (Chi2: P = 0.13; I2 = 43%) (Fig. 3A). Meta-analysis of weight loss included data from six studies of 480 participants; 240 (50.0%) participants were randomized to UF and the remaining 240 randomized to controls. UF was associated with a significant net weight loss as compared to control (MD 1.23, 95% CI 0.03 to 2.44, P = 0.04) with significant, moderate
Fig. 1. Study selection flow diagram.
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
Study [year]
N
UF system
Concomitant HF therapy
Control
Mean age (yr) [% male]
Cr (mg/dL)
LVEF (%)
Follow-up
Agostoni et al. [1993] [15]
36
Gambro System AK 10
Yes
No UF
?
16
Gambro System AK 10
Yes
Agostoni et al. [1995] [17]
42
Gambro System AK 10
Yes
IV furosemide (bolus 160 mg, then 1 mg/min infusion) No UF
UF: 23.8 ± 2.1 Control: 24.1 ± 1.8 ?
180 days
Agostoni et al. [1994] [16]
UF: range 49–69 [88.9] Control: range 50–74 [77.8] UF: 58 ± 1.6 [87.5] Control: 62 ± 2 [87.5]
?
188
Aquadex System 100
No
IV diuretics
Chung et al. [2010] [20,21]
16
?
No
IV furosemide infusion
CUORE [2011] [22–25]
12
Dedyca
?
Conventional management
Hanna et al. [2012] [26]
36
NxStage System One
IV diuretics
Pepi et al. [1993] [27]
24
Gambro System AK10
Stopped except spironolactone (≤25 mg/day) Yes
UF: 27.2 ± 2.8/26.8 ± 2.9b Control: 24.2 ± 1.5/24.6 ± 1.7b UF: 30 (20–52) Control: 35 (25–55) UF: 22 ± 8 Control: 26 ± 11 UF: 39 ± 1 Control: 35 ± 5 UF: 19 ± 9 Control: 18 ± 6
3 months
CARRESS-HF [2012] [18,19]
UF: 58 ± 3/63 ± 3b Control: 60 ± 4/61 ± 4b UF: 69 (61–78) [78] Control: 66 (57–78) [72] UF: 69 ± 14 Control: 74 ± 12 UF: 78.2 ± 2 Control: 72.8 ± 7 UF: 60 ± 9.1 [84.2] Control: 59 ± 15.5 [76]
RAPID-CHF [2005] [28]
40
No
Standard CHF therapy
Rogers et al. [2008] [29]
19
Simple Access Fluid Extraction (SAFE) UFC 100 Console Aquadex System 100
Yes
IV furosemide
ULTRADISCO [2011] [30]
30
PRISMA System
Yes
UNLOAD[2007] [31,32]
200
Aquadex System 100
Yes
IV furosemide infusion (initial dose 250 mg/24 h) IV diuretics
a b c d
No UF
?
UF: 1.90 (1.57–2.37) Control: 2.09 (1.71–2.65) UF: 1.9 ± 0.6 Control: 1.4 ± 0.5 ? UF: 1.6 ± 0.7 Control: 1.7 ± 0.8
UF: 57 ± 5 Control: 56 ± 5 UF: 67.5 [70] Control: 69.5 [70]
?
UF: 64 ± 15 [78] Control: 54 ± 16 [60] UF: 72.4 ± 14.1 [87] Control: 65.8 ± 18.4 [87] UF: 62 ± 15 Control: 63 ± 14
UF: 1.8 ± 0.8 Control: 1.6 ± 0.8 UF: 2.22 ± 0.75 Control: 1.86 ± 0.63 UF: 1.5 ± 0.5 Control: 1.5 ± 0.5
UF: 1.6 Control: 1.8
3 months
60 days 90 days 12 months 90 days
UF: 23.8 ± 8.7 Control: 24.2 ± 9 UF: 69%c Control: 78%c
90 days
?
48 h
UF: 34 ± 19.9 Control: 30 ± 13.4 UF: 71%d Control: 70%d
36 h
30 days
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
Table 1 Characteristics of included studies.a
90 days
Values are mean ± standard deviation or median (interquartile range) unless indicated otherwise. Participants were further stratified into two groups according to whether there was a N10% increase in peak VO2 at 3 months. Percentage of participants with LVEF b 40%. Percentage of participants with LVEF ≤ 40%.
3
4
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
analysis of adverse events was inconclusive since usable data were only available from one study (CARRESS-HF) [18,19]. 3.4. Sensitivity analysis Findings of our sensitivity analyses using the fixed-effect model to test for potential small-study effects were consistent with those of the overall random-effect analyses (Table 2). Sensitivity analyses exploring the potential influence of publication status by excluding data from Chung et al. and CUORE also showed similar results as those of the overall analyses (Table 2); [20–25] the impact of the varied follow-up durations on all-cause mortality was also explored by excluding data from acute studies (Rogers et al. and ULTRADISCO) [29,30], and findings of this sensitivity analysis were found to be comparable to the overall analysis (Table 2). 4. Discussion
Fig. 2. Assessment of risk of bias of included studies. Low risk of bias (green circles), unclear risk of bias (yellow circles) and high risk of bias (red circles). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
statistical heterogeneity among the five studies (Chi [2]: P = 0.03; I [2] = 59%) (Fig. 3B). All-cause deaths occurred in 32 of 329 (9.7%) participants allocated to UF and in 29 of 330 (8.8%) participants allocated to control. There was no significant difference between groups (OR 1.08, 95% CI 0.63 to 1.86, P = 0.77) (Fig. 4A). All-cause rehospitalization was reported in 71 of 203 (35.0%) participants treated with UF and in 74 of 204 (36.3%) participants treated with control. No significant difference was observed between the UF and control groups (OR 0.89, 95% CI 0.39 to 2.00, P = 0.77) (Fig. 4B). 3.3. Secondary outcomes There were no significant differences in the analyses of change in serum creatinine (MD 0.01, 95% CI -0.30 to 0.32; P = 0.94) or unscheduled medical care (OR 1.07, 95% CI 0.31 to 3.69; P = 0.92). Meta-
This meta-analysis of 12 studies shows that UF is effective in reducing volume overload and weight loss, with no benefits in all-cause mortality or rehospitalization. The nonsignificant differences in renal function and incidence of unscheduled medical care are inconclusive due to limited trial data. Safety profile of UF remains to be ascertained due to a lack of adverse event data in the included studies. Currently available RCTs exploring the effects of UF in ADHF are of small size and overall methodological quality was assessed to be low. Although statistical heterogeneity was observed only in the analysis of weight loss (Chi2: P = 0.03; I2 = 59%), we acknowledge the substantial clinical heterogeneity among our included studies, ranging from follow-up duration to outcomes of interest and how the outcomes were defined. Although follow-up duration was found to have minimal impact on our analysis of all-cause mortality, it is important to highlight that our analyses of fluid loss, weight loss and change in creatinine pooled study data that were obtained at varied time points (Table 3). The cut-off point of when these clinically important outcomes are measured is crucial in accurate determination of the efficacy of UF in improving clinical congestion and safety of UF in respect to renal function. We therefore call for the need of consistent cut-off time points of when outcomes are assessed in future studies. Moreover, different measures of renal function were used in the 12 included studies, which greatly contributed to the inconclusive findings of our analyses. In fact, use of serum creatinine levels or changes in serum creatinine (e.g. CARRESS-HF) as outcome measures of renal function among our included studies may not accurately reflect renal status [34]. In clinical practice, serum creatinine clearance and GFR are used as tools for renal function assessment. However, only two of our included studies (CARRESS-HF, Rogers et al.) [18,19,29] described mean changes in GFR and none provided information on baseline or follow-up serum creatinine clearance. The use of accurate and consistent measures for assessing renal function in future clinical trials studying the role of UF in ADHF is essential as it would allow for future analyses to effectively test for the effects of UF on renal function. The significant weight loss and fluid removal as well as lack of effects on serum creatinine associated with the treatment of UF as observed in our analyses are contrary to those reported by the most recent randomized study in the field, CARRESS-HF [18,19]. This randomized trial of 188 patients with ADHF and worsened renal function (defined as an elevated serum creatinine level of N 0.3 mg/dL within 12 weeks before or 10 days after admission for HF) showed that UF as compared to stepped pharmacologic therapy did not have a significant impact on fluid loss or weight loss at 96 h but led to an increase in serum creatinine; however, differences between groups with respect to mortality and rehospitalization were nonsignificant. Findings of our meta-analysis and those of CARRESS-HF illustrate that, despite significant relief of congestion with or without preservation of renal function, UF did not lead to an evident reduction in mortality or rehospitalization rates. The relationship between reduction of congestion, variations in
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
5
Fig. 3. Forest plot of (A) fluid removal; and (B) weight loss.
renal function and clinical outcomes remains unclear. In the Diuretic Optimization Strategies Evaluation (DOSE) trial [35], a significantly higher number of patients with ADHF receiving high-dose diuretic strategy (total daily IV furosemide dose 2.5 times their totally daily oral loop diuretic dose) experienced an significant increase in serum creatinine of N0.3 mg/dL as compared to those receiving low-dose diuretic strategy (23% vs. 14%; P = 0.04), greater weight loss (− 8.7 ± 8.5 lb vs. − 6.8 ± 7.8 lb; P = 0.01) and fluid loss (4899 ± 3479 mL vs. 3575 ± 2635 mL; P = 0.001), and yet the composite end point of mortality, rehospitalization and emergency department visit was comparable between groups (HR 0.83, 95% CI 0.60–1.16; P = 0.28). A single-center observational study by Patarroyo and colleagues showed that [36], in 63 patients with ADHF and worsening renal function and who were
resistant to standard medical therapy, UF significantly reduced body weight at 48-h follow-up (102 ± 25 kg vs. 99 ± 23 kg, P b 0.0001) but there was no change in serum creatinine levels (2.2 ± 0.9 mg/dL vs. 2.4 ± 1 mg/dL, P = 0.12), with somewhat alarming inhospital mortality rate of 30% and overall one-year mortality rate of 70%. Successful treatment of congestion thus cannot simply be translated into renal or clinical improvement, and ongoing research is necessary to determine a risk stratification strategy to carefully identify likely responders who might benefit from UF so to achieve a reasonable balance of congestion relief and clinical improvement. The concept of mechanical removal of excess fluid by UF was first proposed nearly 40 years ago by Silverstein and colleagues in five volume-overloaded patients on chronic hemodialysis [37]. Subsequent
Fig. 4. Forest plot of (A) all-cause mortality; and (B) all-cause rehospitalization.
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
6
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
Table 2 Results of sensitivity analyses of (A) primary outcomes; and (B) secondary outcomes. (A) Net fluid removal (mL) MD (95% CI)
Net weight loss (kg) MD (95% CI)
All-cause mortality OR (95% CI)
All-cause rehospitalization OR (95% CI)
Analytical method Random-effects Fixed-effect
1.28 (0.43, 2.12)⁎ 1.24 (0.69, 1.78)⁎
1.23 (0.03, 2.44)⁎ 1.41 (0.76, 2.06)⁎
1.08 (0.63, 1.86) 1.11 (0.65, 1.88)
0.89 (0.39, 2.00) 0.95 (0.63, 1.44)
Publication status All trials Chung et al. and CUORE excluded
1.28 (0.43, 2.12)⁎ NA
1.23 (0.03, 2.44)⁎ 1.52 (0.11, 2.93)⁎
1.08 (0.63, 1.86) 1.11 (0.64, 1.92)
0.89 (0.39, 2.00) 0.96 (0.40, 2.35)
Follow-up duration Rogers et al. and ULTRADISCO excluded
NA
NA
1.08 (0.63, 1.86)
NA
Analysis
(B) Analysis
Change in serum creatinine (mg/dL) MD (95% CI)
Unscheduled medical care OR (95% CI)
Analytical method Random-effects Fixed-effect
0.01 (−0.30, 0.32) 0.09 (−0.05, 0.22)
1.07 (0.31, 3.69) 0.83 (0.51, 1.34)
Publication status All trials Chung et al. and CUORE excluded
0.01 (−0.30, 0.32) 0.08 (−0.30, 0.46)
1.07 (0.31, 3.69) NA
Abbreviations: CI, confidence interval; MD, mean difference; NA, not applicable; OR, odds ratio * P b 0.05
small observational studies in the 1980s showed early promises of UF in fluid and weight loss, symptomatic improvement with increased response to diuretics in patients with refractory congestive HF [38,39]. However, as illustrated by our literature search, randomized studies of moderate size such as the UNLOAD study were only recently published. During this research gap, much has advanced in the development of different UF systems and physicians as well as patients are faced with an increasing number of options. For example, the FDA-approved Aquadex FlexFlow™ system (Gambro AB, Lund, Sweden) and the Dedyca® system (Bellco, Mirandola, Italy) both utilize a simplistic design and do not require expert management by nephrologists [40]. The Aquadex FlexFlow™ is a portable UF device performed via peripheral venous access whereas the Dedyca® system uses a femoral venous approach. Although a direct head-to-head comparative study to explore the relative effects of the different UF systems is not yet available, data have shown different outcomes associated the use of UF via peripheral vs. femoral access. Liang et al. used the Aquadex FlexFlow™ system in 11 patients with refractory HF who are resistant to diuretics or at risk of resistance Table 3 Cut-off time points of measurement of fluid loss, weight loss and change in creatinine clearance. Study
Time point
Fluid loss CARRESS-HF 2012 Hanna 2012 Rogers 2008 ULTRADISCO 2011 UNLOAD 2007
From randomization to 96 h During intervention period From baseline to 48 h From baseline to 36 h From randomization to 48 h
Weight loss CARRESS-HF 2012 Chung 2010 CUORE (short term) Hanna 2012 Rogers 2008 UNLOAD 2007
From randomization to 96 h ? At discharge During intervention period From baseline to 48 h From randomization to 48 h
Change in serum creatinine CARRESS-HF 2012 CUORE (short term) Rogers 2008
From randomization to 96 h At discharge From baseline to 48 h
with an aim to remove 4 L of fluid per a 8-h UF session, and reported a six-month mortality rate of 55%, 45% of patients experienced a worsened renal function (increase of serum creatinine of N0.3 mg/mL) and 45% of patients required dialysis for persistent diuretic-resistant volume overload and uremic symptoms [41]. In contrast, a study by De Maria and colleagues using the Dedyca® system among 42 patients with congestive HF with LVEF b 40% and diuretic resistance gave a more favorable six-month mortality rate of 26%, only 14% of patients were found with worsened renal function (serum creatinine N3 mg/dL and/or glomerular filtration rate (GFR) b30 mL/min) and only 1 patient (2.4%) required dialysis for persistent diuretic-resistant volume overload and uremic symptoms [42]. The similar baseline characteristics of the enrolled participants of the studies by Liang et al. (age: 70 years; LVEF: 43%; serum creatinine: 2.2 mg/dL; GFR: 38 mL/min; pre-UF furosemide dose: 258 mg/day) and by De Maria et al. (age: 69 ± 13 years; LVEF: 32 ± 5%; serum creatinine: 2 ± 0.6 mg/dL; GFR: 58 ± 27 mL/min; pre-UF furosemide dose: 250 mg/day) and the different clinical outcomes reported suggest that there may be potential device-specific effects in using UF for patients with ADHF. Direct comparison of the effects of UF between the different systems should be evaluated using a wellplanned head-to-head study. However, findings of the aforementioned observational studies illustrate potential influences of treatment regimens (type of system, filtration rate, duration) on the efficacy and safety of UF. It is worth highlighting findings of an interesting study by Costanzo et al., which explored if initiating UF in the early phase of hospitalization would result in euvolemia and early discharge (defined as discharge in ≤3 days) [43]. In a group of 20 patients with ADHF, fluid overload and diuretic resistance/renal insufficiency, UF was initiated within 4.7 ± 3.5 h of hospitalization prior to the administration of IV diuretics. Early UF effectively reduced length of hospital stay (average 3.7 ± 1.8 days), for which 60% of patients were discharged in ≤3 days; quality of life (Minnesota Living With Heart Failure Questionnaire (MLWHFQ)) and global assessment improved from pre-treatment stage (MLWHFQ score: 70 ± 18; global assessment: 5.7 ± 1.3) to 90day follow-up (MLWHFQ score: 51 ± 27; global assessment: 2.5 ± 1.5). Besides the timing of UF initiation, filtration rate has also been raised as an important factor, since aggressive volume removal exceeding the plasma refill rate may lead to intravascular volume depletion,
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
hypotension and renal hypoperfusion [44,45]. It would be of interest and relevance to conduct a subgroup analysis investigating the outcomes associated with different UF strategies. However, information on the UF treatment regimens employed by our included studies is limited. Detailed description of UF regimens is essential to determine the most effective and safest access site (peripheral vs. femoral), treatment duration, volume removal rate and quantity. Furthermore, randomized data on potential adverse events associated with UF are suboptimal, with just two of our included studies (CARRESS-HF [18,19], UNLOAD) [31,32] reporting adverse events and only CARRESS-HF provided event numbers per arm; [18,19] CARRESS-HF showed that the incidence of serious adverse events was significantly higher in the UF arm compared to the pharmacological arm (72% vs. 57%, P = 0.03); this high percentage was mainly attributed to heart and renal failure (33% and 18%, respectively) [18,19]. We thus call for the need of consistent and transparent reporting of complications associated with the treatment of UF in future studies to allow for comprehensive analysis of adverse effects. Three ongoing RCTs comparing diuretics and UF in patients with ADHF were identified in our literature search, all of which are estimated to complete in late 2013. The Aquapheresis Versus Intravenous Diuretics and Hospitalizations for Heart Failure (AVOID-HF) study aims to enroll 810 participants with a primary outcome measure of time to first HF event at 90 days; [46] the REverse WOrsening Renal Function in Decompensated Heart Failure (REWORD-HF) estimates to recruit 183 participants with a primary outcome of changes in dyspnea, weight loss, GFR and brain natriuretic peptide (BNP); [47] and the third study aims to assess myocardial blood flow in an estimated study population of 30 [48]. Results of these ongoing studies are much anticipated and future updated meta-analyses incorporating their data will provide further insights to the precise efficacy and safety profiles of UF in patients with ADHF. In addition, the suboptimal methodological quality demonstrated by the currently available RCTs not only highlights the issue of incomplete reporting of trial methodology but also emphasizes the importance of adequate description of trial design and conduct, especially on the domains of sequence generation and allocation concealment. Conclusions of any meta-analyses rely on the validity and integrity of the included studies. Future meta-analyses should therefore pay particular attention to the assessment of risk of bias in included studies in order to provide better insights to the extent to which the conclusions can be drawn. 4.1. Study limitations Limitations of our meta-analysis include the marked clinical heterogeneity and moderate statistical heterogeneity among the included studies in terms of UF system used, sample size, follow-up duration and reporting of outcomes. The diverse randomized evidence available indicates that findings of our analyses should be interpreted with caution. 5. Conclusions Currently available knowledge on the efficacy and safety of UF in patients with ADHF is inconclusive, where published studies are of small size and reporting of relevant outcome measures is suboptimal. Wellconducted randomized trials with adequate power and carefully selected endpoints are much warranted to enrich the existing body of evidence on the role of UF in patients with refractory HF and resistance to standard diuretic therapy or in whom such therapy is contraindicated. Acknowledgment This study was supported by a research grant from the University Grants Committee of Hong Kong (RGC Collaborative Research Fund 2010/11: CUHK9/CRF/10). We also acknowledge support from the Lui Che Woo Foundation.
7
References [1] Go AS, Mozaffarian D, Roger VL, et al; on behalf of the American Heart Association Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke statistics — 2013 update: a report from the American Heart Association. Circulation 2013;127:e6-245. [2] Lindenfeld J, Albert NM, Boehmer JP, et al. HFSA 2010 comprehensive heart failure practice guideline. J Card Fail 2010;16:e1-194. [3] Onwuanyi A, Taylor M. Acute decompensated heart failure: pathophysiology and treatment. Am J Cardiol 2007;99(6B):25D–30D. [4] Felker GM, Mentz RJ. Diuretics and ultrafiltration in acute decompensated heart failure. J Am Coll Cardiol 2012;59:2145–53. [5] Nieminen MS, Brutsaert D, Dickstein K, et al. EuroHeart Survey Investigators. EuroHeart Failure Survey II (EHFS II): a survey on hospitalized acute heart failure patients: description of population. Eur Heart J 2006;27:2725–36. [6] Adams Jr KF, Fonarow GC, Emerman CL, et al. ADHERE scientific advisory committee and investigators. Characteristics and outcomes of patients hospitalized for heart failure in the United States: rationale, design, and preliminary observations from the first 100,000 cases in the Acute Decompensated Heart Failure National Registry (ADHERE). Am Heart J 2005;149:209–16. [7] McMurray JJ, Adamopoulos S, Anker SD, et al. ESC guidelines for the diagnosis and treatment of acute and chronic heart failure 2012. Eur J Heart Fail 2012;14:803–69. [8] Jessup M, Abraham WT, Casey DE, et al; writing on behalf of the 2005 Guideline Update for the Diagnosis and Management of Chronic Heart Failure in the Adult Writing Committee. 2009 Focused update: ACCF/AHA guidelines for the diagnosis and management of heart failure in adults: a report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines. Circulation 2009;119:1977–2016. [9] Bart BA. Treatment of congestion in congestive heart failure: ultrafiltration is the only rational initial treatment of volume overload in decompensated heart failure. Circ Heart Fail 2009;2:499–504. [10] Wertman BM, Gura V, Schwarz ER. Ultrafiltration for the management of acute decompensated heart failure. J Card Fail 2008;14:754–9. [11] Costanzo MR, Jessup M. Treatment of congestion in heart failure with diuretics and extracorporeal therapies: effects on symptoms, renal function, and prognosis. Heart Fail Rev 2012;17:313–24. [12] Fiaccadori E, Regolisti G, Maggiore U, et al. Ultrafiltration in heart failure. Am Heart J 2011;161:439–49. [13] Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration; 2011 http://handbook.cochrane.org/. [Accessed 25 May 2013]. [14] Moher D, Liberati A, Tetzlaff J, Altman DG; the PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6:e1000097. [15] Agostoni PG, Marenzi GC, Pepi M, et al. Isolated ultrafiltration in moderate congestive heart failure. J Am Coll Cardiol 1993;21:424–31. [16] Agostoni P, Marenzi G, Lauri G, et al. Sustained improvement in functional capacity after removal of body fluid with isolated ultrafiltration in chronic cardiac insufficiency: failure of furosemide to provide the same result. Am J Med 1994;96:191–9. [17] Agostoni PG, Marenzi GC, Sganzerla P, et al. Lung–heart interaction as a substrate for the improvement in exercise capacity after body fluid volume depletion in moderate congestive heart failure. Am J Cardiol 1995;76:793–8. [18] Bart BA, Goldsmith SR, Lee KL, et al; for the Heart Failure Clinical Research Network. Ultrafiltration in decompensated heart failure with cardiorenal syndrome. N Engl J Med 2012;367:2296–304. [19] Bart BA, Goldsmith SR, Lee KL, et al. Cardiorenal rescue study in acute decompensated heart failure: rationale and design of CARRESS-HF, for the Heart Failure Clinical Research Network. J Card Fail 2012;18:176–82. [20] Chung E, Bartone C, Menon S, Mazur W, O'Brien T, Kereiakes D. Target weight guided treatment of acute heart failure using ultrafiltration or usual care: results of a randomized pilot study. Presented at the European Society of Cardiology (ESC) congress; 2010 [28 Aug–01 Sep 2010, Stockholm, Sweden]. [21] Bartone C, Menon SG, Kereiakes DJ, et al. Target weight guided treatment of acute heart failure using ultrafiltration or usual care: results of a randomized pilot study. J Card Fail 2010;16(8S):S106. [22] Cosentino ER, Rinaldi ER, Degli Esposti D, et al. Preliminary report on the effects of ultrafiltration in severe HF refractory to conventional diuretic therapy: the Continuous Ultrafiltration for cOngestive heaRt failurE (CUORE) trial. Eur J Heart Fail Suppl 2011;10:S112. [23] Ferramosca E, Mancini E, Cosentino E, et al. Preliminary report on the effects of ultrafiltration in severe heart failure: the Continuous Ultrafiltration for cOngestive heaRt failurE (CUORE) trial. NDT Plus 2009;2(Suppl. 2):ii427. [24] Muratori M, Agostoni PG, Marenzi G, et al. Preliminary report on the effects of ultrafiltration in severe heart failure: the Continuous Ultrafiltration for cOngestive heaRt failurE (CUORE) trial. Eur J Heart Fail Suppl 2008;7(Suppl. 1):72. [25] Ferramosca E, Cosentino ER, Muratori M, et al; on behalf of the CUORE Trial Investigators. Continuous ultrafiltration for congestive heart failure (CUORE) trial: Risultati preliminari sugli effetti dell'ultrafiltrazione isolata nel trattamento dello scompenso cardiaco severo. G Ital Nefrol 2008;25(S-43):S31. [26] Hanna MA, Tang WH, Teo BW, et al. Extracorporeal ultrafiltration vs. conventional diuretic therapy in advanced decompensated heart failure. Congest Heart Fail 2012;18:54–63. [27] Pepi M, Marenzi GC, Agostoni PG, et al. Sustained cardiac diastolic changes elicited by ultrafiltration in patients with moderate congestive heart failure: pathophysiological correlates. Br Heart J 1993;70:135–40.
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
8
J.S.W. Kwong, C.-M. Yu / International Journal of Cardiology xxx (2014) xxx–xxx
[28] Bart BA, Boyle A, Bank AJ, et al. Ultrafiltration versus usual care for hospitalized patients with heart failure: the relief for acutely fluid-overloaded patients with decompensated congestive heart failure (RAPID-CHF) trial. J Am Coll Cardiol 2005;46:2043–6. [29] Rogers HL, Marshall J, Bock J, et al. A randomized, controlled trial of the renal effects of ultrafiltration as compared to furosemide in patients with acute decompensated heart failure. J Card Fail 2008;14:1–5. [30] Giglioli C, Landi D, Cecchi E, et al. Effects of ULTRAfiltration vs. DIureticS on clinical, biohumoral and haemodynamic variables in patients with deCOmpensated heart failure: the ULTRADISCO study. Eur J Heart Fail 2011;13:337–46. [31] Costanzo MR, Guglin ME, Saltzberg MT, et al; for the UNLOAD Trial Investigators. Ultrafiltration versus intravenous diuretics for patients hospitalized for acute decompensated heart failure. J Am Coll Cardiol 2007;49:675–83. [32] Cleland JG, Coletta AP, Nikitin NP, Clark AL. Clinical trials update from the American College of Cardiology: Darbepoetin alfa, ASTEROID, UNIVERSE, paediatric carvedilol, UNLOAD and ICELAND. Eur J Heart Fail 2006;8:326–9. [33] Stein AC, Mostarda C, Alves B, et al. Ultrafiltration treatment in decompensated heart failure: changes in heart rate variability and survival. Hypertension 2011;58:e64. [34] Kazory A, Ross EA. Ultrafiltration for decompensated heart failure: renal implications. Heart 2009;95:1047–51. [35] Felker GM, Lee KL, Bull DA, et al. NHLBI heart failure clinical research network. Diuretic strategies in patients with acute decompensated heart failure. N Engl J Med 2011;364:797–805. [36] Patarroyo M, Wehbe E, Hanna M, et al. Cardiorenal outcomes after slow continuous ultrafiltration therapy in refractory patients with advanced decompensated heart failure. J Am Coll Cardiol 2012;60:1906–12. [37] Silverstein ME, Ford CA, Lysaght MJ, Henderson LW. Treatment of severe fluid overload by ultrafiltration. N Engl J Med 1974;291:747–51. [38] Simpson IA, Rae AP, Simpson K, et al. Ultrafiltration in the management of refractory congestive heart failure. Br Heart J 1986;55:344–7.
[39] DiLeo M, Pacitti A, Bergerone S, et al. Ultrafiltration in the treatment of refractory congestive heart failure. Clin Cardiol 1988;11:449–52. [40] Wańkowicz Z, Próchnicka A, Olszowska A, Baczyński D, Krzesiński P, Dziuk M. Extracorporeal versus peritoneal ultrafiltration in diuretic-resistant congestive heart failure—a review. Med Sci Monit 2011;17:RA271–81. [41] Liang KV, Hiniker AR, Williams AW, Karon BL, Greene EL, Redfield MM. Use of a novel ultrafiltration device as a treatment strategy for diuretic resistant, refractory heart failure: initial clinical experience in a single center. J Card Fail 2006;12:707–14. [42] De Maria E, Pignatti F, Patrizi G, Benenati PM, Ricci S, Cappelli S. Ultrafiltration for the treatment of diuretic-resistant, recurrent, acute decompensated heart failure: experience in a single center. J Cardiovasc Med 2010;11:599–604. [43] Costanzo MR, Saltzberg M, O'Sullivan J, Sobotka P. Early ultrafiltration in patients with decompensated heart failure and diuretic resistance. J Am Coll Cardiol 2005;46:2047–51. [44] Schroeder KL, Sallustio JE, Ross EA. Continuous haematocrit monitoring during intradialytic hypotension: precipitous decline in plasma refill rates. Nephrol Dial Transplant 2004;19:652–6. [45] Kazory A, Ross EA. Contemporary trends in the pharmacological and extracorporeal management of heart failure: a nephrologic perspective. Circulation 2008;117:975–83. [46] NCT01474200. Study of Heart failure hospitalizations after aquapheresis therapy compared to intravenous (IV) diuretic treatment (AVOID-HF). http://www. clinicaltrials.gov/ct/show/NCT01474200 . [accessed 25 May 2013]. [47] NCT01140399. REWORD-HF REverse WOrsening Renal Function in Decompensated Heart Failure. http://www.clinicaltrials.gov/ct2/show/NCT01140399 . [accessed 25 May 2013]. [48] NCT01457053. Assessment of Coronary Flow Reserve in Heart Failure Patients After Ultrafiltration Versus Diuretics. http://www.clinicaltrials.gov/ct2/show/NCT01457053 . [accessed 25 May 2013].
Please cite this article as: Kwong JSW, Yu C-M, Ultrafiltration for acute decompensated heart failure: A systematic review and meta-analysis of randomized controlled trials, Int J Cardiol (2014), http://dx.doi.org/10.1016/j.ijcard.2014.01.069
