ORIGINAL R ESEARCH AR TICLE
Pentoxifylline in Heart Failure: A Meta-Analysis of Clinical Trials bastien Champion,1 Nathanae €l Lapidus,2 Godefroy Cherie ,3 Vincent Spagnoli,4 Juliette Oliary5 & Se 4 Alain Cohen Solal dicale et Toxicologique, Ho ^pital Lariboisiere, Paris, France 1 R eanimation Me ^pital Saint Antoine, Paris, France 2 Biostatistiques, Ho 3 Pharmacie des batignoles, Paris, France ^pital Lariboisie re, Paris, France 4 Cardiologie, Ho ^pital Lariboisie re, Paris, France 5 Pharmacie, Ho
Keywords heart failure; inflammation; meta-analysis; pentoxifylline; tumor necrosis factor. Correspondence bastien Champion, Re animation Me dicale et Se ^pital Lariboisie re, 2 rue A. Toxicologique, Ho , 75010 Paris, France. Pare Tel.: 00331 49 95 84 42; Fax: 00331 49 95 89 64; E-mail: [email protected]
doi: 10.1111/1755-5922.12076
ABSTRACT Background: Pentoxifylline possess antiinflammatory and rheological properties and has been tested in heart failure (HF). Methods: A comprehensive search was performed from 1980 until July 2013 in PubMed, to identify randomized controlled trials evaluating pentoxifylline versus placebo in HF, to determine impact on mortality. Search strategy is as follows: “Pentoxifylline” AND “heart” AND “trial”. Study selection of six randomized controlled trials evaluating mortality as outcome. Then, we conducted a meta-analysis of randomized controlled trials versus placebo in HF. Determination of Mantel–Haenszel fixed effect and random-effect pooled odds ratios for all-cause mortality and corresponding 95% confidence intervals. Results: Data from a total of 221 patients with LVEF ≤40% from six randomized controlled trials were included in this analysis. Pentoxifylline 1200 mg per day was administered during 6 months, except in one study (administered during 1 month for severe acute HF). The use of pentoxifylline was not significantly associated with a reduction in mortality in HF in individual studies. The pooled data including 221 patients showed a nearly fourfold reduction in mortality (5.4% vs. 18.3%; OR 0.29; CI 0.12–0.74; P < 0.01) with homogenous results (I² 0%). Conclusion: A meta-analysis evaluating pentoxifylline versus placebo in HF suggested a significant nearly fourfold decrease in all-cause mortality in the pentoxifylline group.
Introduction For many years, medical treatment of acute pulmonary edema associated loop diuretics with vasodilators [1]. Recently, mechanical treatment with noninvasive ventilation has been proposed. All these treatments are based on mechanistic concepts of heart failure (HF). HF has been associated with a marked increase in hospital and 1-year mortality occurring in 15% and 35%, respectively [2]. Effective HF prevention may further improve the clinical outcome, and until recently, in the acute phase of acute HF, no medical or mechanical treatment has been proved to decrease this tremendous midterm mortality. Medical treatment of chronic HF including sympathetic and renin–angiotensin–aldosterone neurohumoral blockade carries substantial decrease in mortality [1]. The change in paradigm has proved dramatic lifesaving efficacy in HF. However, introduction of chronic treatment shortly after pulmonary edema is still an issue. Pentoxifylline is a xanthic base with vasodilatatory, antiinflammatory, antiapoptotic, and rheological properties [3]. This drug has already been tested in different pathological conditions such as limb ischemia, acute lung injury, sepsis, cancer, and alcoholic hepatitis. HF that conferred a systemic injury, as proved when
ª 2014 John Wiley & Sons Ltd
cardio-renal or cardio-arterial syndrome occurred, may benefit of new therapeutic target. As the theory of the failing heart moved from the cardio-renal or cardio-circulatory (or hemodynamic) model to the neurohormonal model, the treatment of HF improved. Inflammation is now considered as a major determinant of the disease and a well-known prognosis factor. HF is accompanied by the elevation of the plasma level of ultra-sensible C-reactive protein, tumor necrosis factor alpha, and interleukin 6 and is associated with an increment of pro-apoptotic factors such as Fas/APO-1 [4]. Therefore, pentoxifylline could be useful in the context of HF. Up to now, impact on outcomes of pentoxifylline in HF is still debated. Therefore, we performed an updated meta-analysis of all the randomized controlled trials in which pentoxifylline was compared with placebo to reach a conclusion about the effectiveness of this drug in HF.
Methods Search Strategy We performed a primary search for randomized controlled trials in PubMed. The search terms used were pentoxifylline, heart, and
Cardiovascular Therapeutics 32 (2014) 159–162
159
S. Champion et al.
Pentoxifylline in Heart Failure
trials from 1980 to July 2013. We also searched clinicaltrials.org for published results about trials using pentoxifylline versus placebo in the topics “heart failure” and “heart diseases.” We found 50 citations with PubMed and no additional published study with clinicaltrials.org.
Study Selection Our purpose was the all-cause mortality assessment in randomized controlled trials. Studies were considered if written in English, used a controlled clinical trial design, were conducted in humans, and reported mortality at least 1 month after onset of treatment. We included all modalities of administration, posology, and types of HF: chronic, acutely decompensated HF, and low output or pulmonary edema. We excluded 44 studies of the 50 ones returned by the PubMed search.
cation or patients’ overlap was acknowledged in the six trials conducted in Baragwanath Hospital in Johannesburg [5–7,9,10]. The last study included in our meta-analysis was performed by a different center in a different country [12]. All the patients had decreased left ventricular ejection fraction (LVEF) (≤40%). They were treated with beta-blockers and angiotensin-converting enzyme inhibitors, except for the two first studies published before 2001 [5,6]. Mortality was assessed at 6 months for five studies evaluating decompensated “chronic” HF [5–7,10,12] and at 1 month for one study in severe acute HF syndrome [9]. No study found a significant difference in mortality with pentoxifylline versus control in HF, but the pooled data including 221 patients showed a nearly fourfold reduction in mortality (5.4% vs. 18.3%; OR 0.29; CI 0.12–0.74; P < 0.01). The results of the hospital mortality meta-analysis of all the randomized studies are shown in Figure 1 and Table 1 [5–7,9,10,12].
Statistics We calculated Mantel–Haenszel fixed effect and random-effect pooled odds ratios and corresponding 95% CI for pentoxifylline treatment versus placebo. Statistical heterogeneity among study results was investigated using the I2 statistics. We reported the association with mortality with the random-effects odds ratio. Statistical analyses were conducted with R version 2.15.1 (R Development Core Team 2011; R Foundation for Statistical Computing, Vienna, Austria). As mortality was assessed at 6 months for five studies (decompensated “chronic” HF) and at 1 month for one study (severe acute HF syndrome), we conducted a sensitivity analysis after selecting only those with 6-month mortality assessment.
Results From 1998 to 2004, we identified a total of seven studies evaluating the effects of pentoxifylline on hospital mortality in HF. The first six studies [5–10] are from a single center with a meta-analysis already performed for the five randomized studies [11]. One of these studies was excluded because of its before/after design [8]. This study showed a significant decrease in hospital mortality in peripartum cardiomyopathy over time after introduction of pentoxifylline in the therapeutic armatorium. No duplicate publi-
Discussion For the past 15 years, some studies have evaluated pentoxifylline in HF. Indeed, HF is associated with systemic inflammation, and pentoxifylline could limit this adverse effect. Randomized controlled trials have already been performed to compare pentoxifylline to placebo without showing a significant impact on mortality. In the above-mentioned meta-analysis [11], the mortality did not significantly differ in patients receiving pentoxifylline when compared to patient receiving placebo, but showed a significant improvement in dyspnea and LVEF in patients receiving pentoxifylline. For this reason, we decided to conduct another metaanalysis of randomized controlled trials evaluating pentoxifylline, after exclusion of this last study. The first major result of our meta-analysis was that pentoxifylline was associated with a fourfold decrease in mortality in a heterogeneous group of HF patients but with homogenous results (I² 0%). Secondly, we were unable to determine a subtype of HF that might benefit from pentoxifylline ever dilated, ischemic or with marked blood inflammation. The last subset of patients with elevated C-reactive protein could be “pentoxifylline responders” and might benefit of this immunomodulatory therapy [13]. However, all the studies evaluated patients with HF with reduced
Figure 1 Prognosis meta-analysis of the six published randomized controlled trials evaluating mortality outcome of pentoxifylline in HF.
160
Cardiovascular Therapeutics 32 (2014) 159–162
ª 2014 John Wiley & Sons Ltd
S. Champion et al.
Pentoxifylline in Heart Failure
Table 1 Design and main results of included studies
References
Name of the first author
Year of publication
Delay of mortality assessment (months)
[5] [6] [7] [9] [10] [12]
Sliwa Skudicky Skudicky Sliwa Sliwa Bahrmann
1998 2000 2001 2002 2004 2004
6 6 6 1 6 6
CM Dilated Dilated Dilated Dilated Ischemic Ischemic, hypertensive, and dilated
Mortality of the pentoxifylline group (%)
Mortality of the placebo group (%)
0/14 2/25 2/20 1/9 1/21 0/23
4/14 6/24 3/20 2/9 4/18 1/24
(0) (8) (10) (11) (5) (0)
(29) (25) (15) (22) (22) (4)
CM, cardiomyopathy.
LVEF. Then, pentoxifylline was effective and well tolerated in stable or decompensated HF. It was effective in patients’ na€ıve from or receiving beta-blockers or angiotensin-converting enzyme inhibitors. Our meta-analysis included five of the six trials conducted in Baragwanath Hospital in Johannesburg [5–7,9,10]. Of the six trials, five were controlled and randomized and involved dilated and ischemic cardiomyopathies. Some of them acknowledged industrial sponsor [10] and all investigated a 400 mg three times a day pentoxifylline administration [5–10]. Mortality assessment is blinded from randomization at 6 months except for the most severe class four patients evaluated at 1 month [9]. To accord for this different follow-up, we carried out a sensitivity analysis in which we discarded this last study with similar results. The last randomized controlled study conducted by another team evaluated efficacy of a 600 mg twice daily of pentoxifylline for 6 months in chronic HF including ischemic, dilated, or hypertensive cardiomyopathy [12]. These administration modalities might have influenced outcome because pentoxifylline has a particular pharmacokinetics in patients with HF [14]. One more randomized controlled study evaluating pentoxifylline versus placebo in 45 patients with left ventricular dysfunction had started in 2011, but results are still waited for July 2014. Nevertheless, our study adds evidence of the effectiveness of pentoxifylline in midterm mortality of HF and calls for new randomized controlled trials in different subsets of HF. As a result, pentoxifylline could be either an “add-on” therapy of HF by immunomodulatory effects or a transition therapy after acute event waiting treatment with full purpose dose of beta-blockers and angiotensin-converting enzyme inhibitors [15].
Limitations It is hard to draw firm conclusion about the real effectiveness of pentoxifylline in decreasing by fourfold the mortality in HF as this result is surprisingly optimist and our meta-analysis has limita-
References 1. McMurray JJ, Adamopoulos S, Anker SD, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The
tions. First, our analysis included only six randomized studies accounting for 221 patients. None of the studies including in this meta-analysis were powered to detect differences in mortality. Even when pooling all studies together, the sample size remains very small. Our findings only suggest a possible reduction in mortality with pentoxifylline. Definitively, further patients are needed before advocating such tremendous reduction in mortality. Second, even if the results seem homogenous, different types of cardiomyopathy such as dilated, hypertensive or ischemic, and even low output syndrome were included and with many inclusion criteria and different lengths of pentoxifylline administration. Furthermore, most trials were performed by a single institution. Basically, more randomized controlled trials are needed before liberal use of pentoxifylline in HF.
Conclusion A meta-analysis of all trials evaluating pentoxifylline versus control in HF suggested a significant fourfold decrease in all-cause mortality in the pentoxifylline group.
Disclosure of Funding None.
Author Contributions Sebastien Champion contributed in design, writing, and drafting. Nathana€el Lapidus contributed in statistics. Godefroy Cherie carried out data collection. Vincent Spagnoli performed design and writing. Juliette Oliary performed design and writing. Alain Cohen Solal contributed in critical revision of article.
Conflict of Interest The authors declare no conflict of interest.
European Society of Cardiology. Developed in
States: rationale, design, and preliminary
collaboration with the Heart Failure Association
observations from the first 100,000 cases in the
(HFA) of the ESC. Eur Heart J 2012;33:1787–
Acute Decompensated Heart Failure National
1847.
Registry (ADHERE). Am Heart J 2005;149:209–
2. Adams KF Jr, Fonarow GC, Emerman CL, et al.
Task Force for the Diagnosis and Treatment of
Characteristics and outcomes of patients
Acute and Chronic Heart Failure 2012 of the
hospitalized for heart failure in the United
ª 2014 John Wiley & Sons Ltd
16. 3. Zhang M, Xu YJ, Mengi SA, Arneja AS, Dhalla NS. Therapeutic potentials of pentoxifylline for
Cardiovascular Therapeutics 32 (2014) 159–162
161
S. Champion et al.
Pentoxifylline in Heart Failure
treatment of cardiovascular diseases. Exp Clin
randomized study. Circulation 2001;103:1083–
Cardiol 2004;9:103–111.
1088.
4. Shaw SM, Shah MK, Williams SG, Fildes JE.
8. Sliwa K, Skudicky D, Candy G, Bergemann A,
12. Bahrmann P, Hengst UM, Richartz BM, Figulla HR. Pentoxifylline in ischemic, hypertensive and idiopathic-dilated cardiomyopathy: effects
Immunological mechanisms of pentoxifylline in
Hopley M, Sareli P. The addition of
chronic heart failure. Eur J Heart Fail 2009;
pentoxifylline to conventional therapy improves
cytokines and symptoms. Eur J Heart Fail
11:113–118.
outcome in patients with peripartum
2004;6:195–201.
5. Sliwa K, Skudicky D, Candy G, Wisenbaugh T, Sareli P. Randomised investigation of effects of pentoxifylline on left-ventricular performance
cardiomyopathy. Eur J Heart Fail 2002;4:305– 309. 9. Sliwa K, Woodiwiss A, Candy G, et al. Effects of
in idiopathic dilated cardiomyopathy. Lancet
pentoxifylline on cytokine profiles and left
1998;351:1091–1093.
ventricular performance in patients with
6. Skudicky D, Sliwa K, Bergemann A, Candy G,
decompensated congestive heart failure
on left-ventricular function, inflammatory
13. Sliwa K, Woodiwiss A, Libhaber E, Zhanje F, Libhaber C, Motara R, Essop R. C-reactive protein predicts response to pentoxifylline in patients with idiopathic dilated cardiomyopathy. Eur J Heart Fail 2004;6:731–734. 14. Nisi A, Panfili M, De Rosa G, Boffa G, Groppa F,
Sareli P. Reduction in Fas/APO-1 plasma
secondary to idiopathic dilated cardiomyopathy.
Gusella M, Padrini R. Pharmacokinetics of
concentrations correlates with improvement in
Am J Cardiol 2002;90:1118–1122.
pentoxifylline and its main metabolites in
left ventricular function in patients with
10. Sliwa K, Woodiwiss A, Kone VN, et al. Therapy
idiopathic dilated cardiomyopathy treated with
of ischemic cardiomyopathy with the
pentoxifylline. Heart 2000;84:438–439.
immunomodulating agent pentoxifylline: results
7. Skudicky D, Bergemann A, Sliwa K, Candy G, Sareli P. Beneficial effects of pentoxifylline in patients with idiopathic dilated cardiomyopathy
of a randomized study. Circulation 2004;109: 750–755. 11. Batchelder K, Mayosi BM. Pentoxifylline for
treated with angiotensin-converting enzyme
heart failure: a systematic review. S Afr Med J
inhibitors and carvedilol: results of a
2005;95:171–175.
162
Cardiovascular Therapeutics 32 (2014) 159–162
patients with different degrees of heart failure following a single dose of a modified-release formulation. J Clin Pharmacol 2013;53:51–57. 15. Hofmann U, Frantz S. How can we cure a heart “in flame”? A translational view on inflammation in heart failure. Basic Res Cardiol 2013;108:356.
ª 2014 John Wiley & Sons Ltd
Pentoxifylline in Heart Failure: A Meta-Analysis of Clinical Trials bastien Champion,1 Nathanae €l Lapidus,2 Godefroy Cherie ,3 Vincent Spagnoli,4 Juliette Oliary5 & Se 4 Alain Cohen Solal dicale et Toxicologique, Ho ^pital Lariboisiere, Paris, France 1 R eanimation Me ^pital Saint Antoine, Paris, France 2 Biostatistiques, Ho 3 Pharmacie des batignoles, Paris, France ^pital Lariboisie re, Paris, France 4 Cardiologie, Ho ^pital Lariboisie re, Paris, France 5 Pharmacie, Ho
Keywords heart failure; inflammation; meta-analysis; pentoxifylline; tumor necrosis factor. Correspondence bastien Champion, Re animation Me dicale et Se ^pital Lariboisie re, 2 rue A. Toxicologique, Ho , 75010 Paris, France. Pare Tel.: 00331 49 95 84 42; Fax: 00331 49 95 89 64; E-mail: [email protected]
doi: 10.1111/1755-5922.12076
ABSTRACT Background: Pentoxifylline possess antiinflammatory and rheological properties and has been tested in heart failure (HF). Methods: A comprehensive search was performed from 1980 until July 2013 in PubMed, to identify randomized controlled trials evaluating pentoxifylline versus placebo in HF, to determine impact on mortality. Search strategy is as follows: “Pentoxifylline” AND “heart” AND “trial”. Study selection of six randomized controlled trials evaluating mortality as outcome. Then, we conducted a meta-analysis of randomized controlled trials versus placebo in HF. Determination of Mantel–Haenszel fixed effect and random-effect pooled odds ratios for all-cause mortality and corresponding 95% confidence intervals. Results: Data from a total of 221 patients with LVEF ≤40% from six randomized controlled trials were included in this analysis. Pentoxifylline 1200 mg per day was administered during 6 months, except in one study (administered during 1 month for severe acute HF). The use of pentoxifylline was not significantly associated with a reduction in mortality in HF in individual studies. The pooled data including 221 patients showed a nearly fourfold reduction in mortality (5.4% vs. 18.3%; OR 0.29; CI 0.12–0.74; P < 0.01) with homogenous results (I² 0%). Conclusion: A meta-analysis evaluating pentoxifylline versus placebo in HF suggested a significant nearly fourfold decrease in all-cause mortality in the pentoxifylline group.
Introduction For many years, medical treatment of acute pulmonary edema associated loop diuretics with vasodilators [1]. Recently, mechanical treatment with noninvasive ventilation has been proposed. All these treatments are based on mechanistic concepts of heart failure (HF). HF has been associated with a marked increase in hospital and 1-year mortality occurring in 15% and 35%, respectively [2]. Effective HF prevention may further improve the clinical outcome, and until recently, in the acute phase of acute HF, no medical or mechanical treatment has been proved to decrease this tremendous midterm mortality. Medical treatment of chronic HF including sympathetic and renin–angiotensin–aldosterone neurohumoral blockade carries substantial decrease in mortality [1]. The change in paradigm has proved dramatic lifesaving efficacy in HF. However, introduction of chronic treatment shortly after pulmonary edema is still an issue. Pentoxifylline is a xanthic base with vasodilatatory, antiinflammatory, antiapoptotic, and rheological properties [3]. This drug has already been tested in different pathological conditions such as limb ischemia, acute lung injury, sepsis, cancer, and alcoholic hepatitis. HF that conferred a systemic injury, as proved when
ª 2014 John Wiley & Sons Ltd
cardio-renal or cardio-arterial syndrome occurred, may benefit of new therapeutic target. As the theory of the failing heart moved from the cardio-renal or cardio-circulatory (or hemodynamic) model to the neurohormonal model, the treatment of HF improved. Inflammation is now considered as a major determinant of the disease and a well-known prognosis factor. HF is accompanied by the elevation of the plasma level of ultra-sensible C-reactive protein, tumor necrosis factor alpha, and interleukin 6 and is associated with an increment of pro-apoptotic factors such as Fas/APO-1 [4]. Therefore, pentoxifylline could be useful in the context of HF. Up to now, impact on outcomes of pentoxifylline in HF is still debated. Therefore, we performed an updated meta-analysis of all the randomized controlled trials in which pentoxifylline was compared with placebo to reach a conclusion about the effectiveness of this drug in HF.
Methods Search Strategy We performed a primary search for randomized controlled trials in PubMed. The search terms used were pentoxifylline, heart, and
Cardiovascular Therapeutics 32 (2014) 159–162
159
S. Champion et al.
Pentoxifylline in Heart Failure
trials from 1980 to July 2013. We also searched clinicaltrials.org for published results about trials using pentoxifylline versus placebo in the topics “heart failure” and “heart diseases.” We found 50 citations with PubMed and no additional published study with clinicaltrials.org.
Study Selection Our purpose was the all-cause mortality assessment in randomized controlled trials. Studies were considered if written in English, used a controlled clinical trial design, were conducted in humans, and reported mortality at least 1 month after onset of treatment. We included all modalities of administration, posology, and types of HF: chronic, acutely decompensated HF, and low output or pulmonary edema. We excluded 44 studies of the 50 ones returned by the PubMed search.
cation or patients’ overlap was acknowledged in the six trials conducted in Baragwanath Hospital in Johannesburg [5–7,9,10]. The last study included in our meta-analysis was performed by a different center in a different country [12]. All the patients had decreased left ventricular ejection fraction (LVEF) (≤40%). They were treated with beta-blockers and angiotensin-converting enzyme inhibitors, except for the two first studies published before 2001 [5,6]. Mortality was assessed at 6 months for five studies evaluating decompensated “chronic” HF [5–7,10,12] and at 1 month for one study in severe acute HF syndrome [9]. No study found a significant difference in mortality with pentoxifylline versus control in HF, but the pooled data including 221 patients showed a nearly fourfold reduction in mortality (5.4% vs. 18.3%; OR 0.29; CI 0.12–0.74; P < 0.01). The results of the hospital mortality meta-analysis of all the randomized studies are shown in Figure 1 and Table 1 [5–7,9,10,12].
Statistics We calculated Mantel–Haenszel fixed effect and random-effect pooled odds ratios and corresponding 95% CI for pentoxifylline treatment versus placebo. Statistical heterogeneity among study results was investigated using the I2 statistics. We reported the association with mortality with the random-effects odds ratio. Statistical analyses were conducted with R version 2.15.1 (R Development Core Team 2011; R Foundation for Statistical Computing, Vienna, Austria). As mortality was assessed at 6 months for five studies (decompensated “chronic” HF) and at 1 month for one study (severe acute HF syndrome), we conducted a sensitivity analysis after selecting only those with 6-month mortality assessment.
Results From 1998 to 2004, we identified a total of seven studies evaluating the effects of pentoxifylline on hospital mortality in HF. The first six studies [5–10] are from a single center with a meta-analysis already performed for the five randomized studies [11]. One of these studies was excluded because of its before/after design [8]. This study showed a significant decrease in hospital mortality in peripartum cardiomyopathy over time after introduction of pentoxifylline in the therapeutic armatorium. No duplicate publi-
Discussion For the past 15 years, some studies have evaluated pentoxifylline in HF. Indeed, HF is associated with systemic inflammation, and pentoxifylline could limit this adverse effect. Randomized controlled trials have already been performed to compare pentoxifylline to placebo without showing a significant impact on mortality. In the above-mentioned meta-analysis [11], the mortality did not significantly differ in patients receiving pentoxifylline when compared to patient receiving placebo, but showed a significant improvement in dyspnea and LVEF in patients receiving pentoxifylline. For this reason, we decided to conduct another metaanalysis of randomized controlled trials evaluating pentoxifylline, after exclusion of this last study. The first major result of our meta-analysis was that pentoxifylline was associated with a fourfold decrease in mortality in a heterogeneous group of HF patients but with homogenous results (I² 0%). Secondly, we were unable to determine a subtype of HF that might benefit from pentoxifylline ever dilated, ischemic or with marked blood inflammation. The last subset of patients with elevated C-reactive protein could be “pentoxifylline responders” and might benefit of this immunomodulatory therapy [13]. However, all the studies evaluated patients with HF with reduced
Figure 1 Prognosis meta-analysis of the six published randomized controlled trials evaluating mortality outcome of pentoxifylline in HF.
160
Cardiovascular Therapeutics 32 (2014) 159–162
ª 2014 John Wiley & Sons Ltd
S. Champion et al.
Pentoxifylline in Heart Failure
Table 1 Design and main results of included studies
References
Name of the first author
Year of publication
Delay of mortality assessment (months)
[5] [6] [7] [9] [10] [12]
Sliwa Skudicky Skudicky Sliwa Sliwa Bahrmann
1998 2000 2001 2002 2004 2004
6 6 6 1 6 6
CM Dilated Dilated Dilated Dilated Ischemic Ischemic, hypertensive, and dilated
Mortality of the pentoxifylline group (%)
Mortality of the placebo group (%)
0/14 2/25 2/20 1/9 1/21 0/23
4/14 6/24 3/20 2/9 4/18 1/24
(0) (8) (10) (11) (5) (0)
(29) (25) (15) (22) (22) (4)
CM, cardiomyopathy.
LVEF. Then, pentoxifylline was effective and well tolerated in stable or decompensated HF. It was effective in patients’ na€ıve from or receiving beta-blockers or angiotensin-converting enzyme inhibitors. Our meta-analysis included five of the six trials conducted in Baragwanath Hospital in Johannesburg [5–7,9,10]. Of the six trials, five were controlled and randomized and involved dilated and ischemic cardiomyopathies. Some of them acknowledged industrial sponsor [10] and all investigated a 400 mg three times a day pentoxifylline administration [5–10]. Mortality assessment is blinded from randomization at 6 months except for the most severe class four patients evaluated at 1 month [9]. To accord for this different follow-up, we carried out a sensitivity analysis in which we discarded this last study with similar results. The last randomized controlled study conducted by another team evaluated efficacy of a 600 mg twice daily of pentoxifylline for 6 months in chronic HF including ischemic, dilated, or hypertensive cardiomyopathy [12]. These administration modalities might have influenced outcome because pentoxifylline has a particular pharmacokinetics in patients with HF [14]. One more randomized controlled study evaluating pentoxifylline versus placebo in 45 patients with left ventricular dysfunction had started in 2011, but results are still waited for July 2014. Nevertheless, our study adds evidence of the effectiveness of pentoxifylline in midterm mortality of HF and calls for new randomized controlled trials in different subsets of HF. As a result, pentoxifylline could be either an “add-on” therapy of HF by immunomodulatory effects or a transition therapy after acute event waiting treatment with full purpose dose of beta-blockers and angiotensin-converting enzyme inhibitors [15].
Limitations It is hard to draw firm conclusion about the real effectiveness of pentoxifylline in decreasing by fourfold the mortality in HF as this result is surprisingly optimist and our meta-analysis has limita-
References 1. McMurray JJ, Adamopoulos S, Anker SD, et al. ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure 2012: The
tions. First, our analysis included only six randomized studies accounting for 221 patients. None of the studies including in this meta-analysis were powered to detect differences in mortality. Even when pooling all studies together, the sample size remains very small. Our findings only suggest a possible reduction in mortality with pentoxifylline. Definitively, further patients are needed before advocating such tremendous reduction in mortality. Second, even if the results seem homogenous, different types of cardiomyopathy such as dilated, hypertensive or ischemic, and even low output syndrome were included and with many inclusion criteria and different lengths of pentoxifylline administration. Furthermore, most trials were performed by a single institution. Basically, more randomized controlled trials are needed before liberal use of pentoxifylline in HF.
Conclusion A meta-analysis of all trials evaluating pentoxifylline versus control in HF suggested a significant fourfold decrease in all-cause mortality in the pentoxifylline group.
Disclosure of Funding None.
Author Contributions Sebastien Champion contributed in design, writing, and drafting. Nathana€el Lapidus contributed in statistics. Godefroy Cherie carried out data collection. Vincent Spagnoli performed design and writing. Juliette Oliary performed design and writing. Alain Cohen Solal contributed in critical revision of article.
Conflict of Interest The authors declare no conflict of interest.
European Society of Cardiology. Developed in
States: rationale, design, and preliminary
collaboration with the Heart Failure Association
observations from the first 100,000 cases in the
(HFA) of the ESC. Eur Heart J 2012;33:1787–
Acute Decompensated Heart Failure National
1847.
Registry (ADHERE). Am Heart J 2005;149:209–
2. Adams KF Jr, Fonarow GC, Emerman CL, et al.
Task Force for the Diagnosis and Treatment of
Characteristics and outcomes of patients
Acute and Chronic Heart Failure 2012 of the
hospitalized for heart failure in the United
ª 2014 John Wiley & Sons Ltd
16. 3. Zhang M, Xu YJ, Mengi SA, Arneja AS, Dhalla NS. Therapeutic potentials of pentoxifylline for
Cardiovascular Therapeutics 32 (2014) 159–162
161
S. Champion et al.
Pentoxifylline in Heart Failure
treatment of cardiovascular diseases. Exp Clin
randomized study. Circulation 2001;103:1083–
Cardiol 2004;9:103–111.
1088.
4. Shaw SM, Shah MK, Williams SG, Fildes JE.
8. Sliwa K, Skudicky D, Candy G, Bergemann A,
12. Bahrmann P, Hengst UM, Richartz BM, Figulla HR. Pentoxifylline in ischemic, hypertensive and idiopathic-dilated cardiomyopathy: effects
Immunological mechanisms of pentoxifylline in
Hopley M, Sareli P. The addition of
chronic heart failure. Eur J Heart Fail 2009;
pentoxifylline to conventional therapy improves
cytokines and symptoms. Eur J Heart Fail
11:113–118.
outcome in patients with peripartum
2004;6:195–201.
5. Sliwa K, Skudicky D, Candy G, Wisenbaugh T, Sareli P. Randomised investigation of effects of pentoxifylline on left-ventricular performance
cardiomyopathy. Eur J Heart Fail 2002;4:305– 309. 9. Sliwa K, Woodiwiss A, Candy G, et al. Effects of
in idiopathic dilated cardiomyopathy. Lancet
pentoxifylline on cytokine profiles and left
1998;351:1091–1093.
ventricular performance in patients with
6. Skudicky D, Sliwa K, Bergemann A, Candy G,
decompensated congestive heart failure
on left-ventricular function, inflammatory
13. Sliwa K, Woodiwiss A, Libhaber E, Zhanje F, Libhaber C, Motara R, Essop R. C-reactive protein predicts response to pentoxifylline in patients with idiopathic dilated cardiomyopathy. Eur J Heart Fail 2004;6:731–734. 14. Nisi A, Panfili M, De Rosa G, Boffa G, Groppa F,
Sareli P. Reduction in Fas/APO-1 plasma
secondary to idiopathic dilated cardiomyopathy.
Gusella M, Padrini R. Pharmacokinetics of
concentrations correlates with improvement in
Am J Cardiol 2002;90:1118–1122.
pentoxifylline and its main metabolites in
left ventricular function in patients with
10. Sliwa K, Woodiwiss A, Kone VN, et al. Therapy
idiopathic dilated cardiomyopathy treated with
of ischemic cardiomyopathy with the
pentoxifylline. Heart 2000;84:438–439.
immunomodulating agent pentoxifylline: results
7. Skudicky D, Bergemann A, Sliwa K, Candy G, Sareli P. Beneficial effects of pentoxifylline in patients with idiopathic dilated cardiomyopathy
of a randomized study. Circulation 2004;109: 750–755. 11. Batchelder K, Mayosi BM. Pentoxifylline for
treated with angiotensin-converting enzyme
heart failure: a systematic review. S Afr Med J
inhibitors and carvedilol: results of a
2005;95:171–175.
162
Cardiovascular Therapeutics 32 (2014) 159–162
patients with different degrees of heart failure following a single dose of a modified-release formulation. J Clin Pharmacol 2013;53:51–57. 15. Hofmann U, Frantz S. How can we cure a heart “in flame”? A translational view on inflammation in heart failure. Basic Res Cardiol 2013;108:356.
ª 2014 John Wiley & Sons Ltd
