Accepted Manuscript Effect of Albumin in Cirrhotic Patients with Infection Other Than Spontaneous Bacterial Peritonitis. A Randomized Trial Thierry Thévenot, Christophe Bureau, Frédéric Oberti, Rodolphe Anty, Alexandre Louvet, Aurélie Plessier, Marika Rudler, Alexandra Heurgué-Berlot, Isabelle Rosa, Nathalie Talbodec, Thong Dao, Violaine Ozenne, Nicolas Carbonell, Xavier Causse, Odile Goria, Anne Minello, Victor De Ledinghen, Roland Amathieu, Hélène Barraud, Eric Nguyen-Khac, Claire Becker, Thierry Paupard, Danielle Botta-Fridlung, Naceur Abdelli, François Guillemot, Elisabeth Monnet, Vincent Di Martino PII: DOI: Reference:
S0168-8278(14)00862-9 http://dx.doi.org/10.1016/j.jhep.2014.11.017 JHEPAT 5438
To appear in:
Journal of Hepatology
Received Date: Revised Date: Accepted Date:
13 August 2014 5 November 2014 6 November 2014
Please cite this article as: Thévenot, T., Bureau, C., Oberti, F., Anty, R., Louvet, A., Plessier, A., Rudler, M., Heurgué-Berlot, A., Rosa, I., Talbodec, N., Dao, T., Ozenne, V., Carbonell, N., Causse, X., Goria, O., Minello, A., De Ledinghen, V., Amathieu, R., Barraud, H., Nguyen-Khac, E., Becker, C., Paupard, T., Botta-Fridlung, D., Abdelli, N., Guillemot, F., Monnet, E., Di Martino, V., Effect of Albumin in Cirrhotic Patients with Infection Other Than Spontaneous Bacterial Peritonitis. A Randomized Trial, Journal of Hepatology (2014), doi: http://dx.doi.org/ 10.1016/j.jhep.2014.11.017
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Page 1, Thévenot T., et al. EFFECT OF ALBUMIN IN CIRRHOTIC PATIENTS WITH INFECTION OTHER THAN SPONTANEOUS BACTERIAL PERITONITIS. A RANDOMIZED TRIAL. Thierry Thévenot (1), Christophe Bureau (2), Frédéric Oberti (3), Rodolphe Anty (4), Alexandre Louvet (5), Aurélie Plessier (6), Marika Rudler (7), Alexandra Heurgué-Berlot (8),Isabelle Rosa (9), Nathalie Talbodec (10), Thong Dao (11), Violaine Ozenne (12), Nicolas Carbonell (13), Xavier Causse (14), Odile Goria (15), Anne Minello (16), Victor De Ledinghen (17), Roland Amathieu (18), Hélène Barraud(19), Eric Nguyen-Khac (20), Claire Becker (21), Thierry Paupard (22), Danielle Botta-Fridlung (23), Naceur Abdelli (24), François Guillemot (25), Elisabeth Monnet (1), Vincent Di Martino (1). d'Hépatologie, hôpital Jean Minjoz, 25000 Besançon, (2)Service d’HépatoGastroentérologie, hôpital Purpan, clinique Dieulafoy, 31059 Toulouse, (3)Service d’ d’HépatoGastroentérologie, hôpital d’Angers, 4 rue Larrey 49100 Angers, (4)Service d’HépatoGastroentérologie, hôpital Archet, rue St Antoine Ginestier, 06200 Nice, (5)Service des Maladies de l’Appareil Digestif, CHRU de Lille, rue M. Polonovski, 59037 Lille cedex (6)Inserm U-773, Service d’Hépatologie, hôpital Beaujon, 100 boulevard du Général Leclerc, 92118 Clichy cédex, (7)Service d’Hépato-Gastroentérologie, hôpital Pitié-Salpétrière, 47-83 boulevard de l’Hôpital, 75651 Paris Cedex 13, (8)Service d’Hépato-Gastroentérologie, hôpital Robert Debré, 51092 Reims cedex, (9)Service d’Hépato-Gastroentérologie, CHIC de Créteil, 40 avenue de Verdun, 94010 Créteil cedex, (10)Service d’Hépato-Gastroentérologie, centre hospitalier, 135 rue du Président Coty, 59 200 Tourcoing, (11)Inserm U-1075, Service d’Hépato-Gastroentérologie, hôpital de Caen, avenue de la Côte de Nacre, 14000 Caen, (12)Service d’Hépato-Gastroentérologie, hôpital Lariboisière, 2 rue AmbroiseParé, 75010 Paris, (13)Service d’Hépato-Gastroentérologie, hôpital Saint-Antoine, 184 rue du Fbg St Antoine, 75571 Paris cedex 12, (14)Service d’Hépato-Gastroentérologie, hôpital de la Source, BP 6709, 45067 Orléans cedex 12, (15)Service d’Hépato-Gastroentérologie, hôpital de Rouen, 1 Rue Germont, 76000 Rouen, (16)Service d’Hépato-Gastroentérologie, hôpital du Bocage, BP 1542, 21034 Dijon cedex, (17)Service d’Hépato-Gastroentérologie, hôpital du haut Levêque, 33604 Pessac Cedex, (18)Service d’Hépato-Gastroentérologie, hôpital Jean Verdier, av. du 14 juillet, 93143 Bondy cedex, (19)Service d’Hépato-Gastroentérologie, hôpital de Brabois, rue du Morvan, 54511 Vandoeuvre-lesNancy cedex, (20)Service d’Hépato-Gastroentérologie, hôpital d’Amiens, 2 Place Victor Pauchet, 80080 Amiens, (21)Service d’Hépato-Gastroentérologie, hôpital de Lens, 99 route de la Bassée, SP-8, 62307 Lens cedex, (22)Service d’Hépato-Gastroentérologie, hôpital de Dunkerque, 130 avenue Louis Herbeaux, 59385 Dunkerque cedex 1, (23)Service d’Hépato-Gastroentérologie, hôpital de la Conception, 147 Bd Baille, 13005 Marseille, (24)Service d’Hépato-Gastroentérologie, hôpital de Chalons-en-Champagne, 51 rue du commandant Derrien, 51005 Chalons-en-Champagne cedex, (25)Service d’Hépato-Gastroentérologie, hôpital de Roubaix, 11-17 Boulevard Lacordaire, 59100 Roubaix. ( 1)Service
Study concept and design : Thevenot T, Di Martino V and Monnet E; Acquisition of data : Thévenot T, Bureau C, Oberti F, Anty R, Louvet A, Plessier A, Rudler M, Heurgué-Berlot A, Rosa I, Talbodec N, Dao T, Ozenne V, Carbonell N, Causse X, Goria O, Minello A, De Ledinghen V, Amathieu R, Barraud H, Nguyen-Khac E, Becker C, Paupard T, Botta-Fridlung D, Abdelli N, Guillemot F; Analysis, interpretation of data and statistical analysis : Thevenot T and Monnet E; drafting of the manuscript and obtained funding : Thevenot T; Technical support: Leroux F, Muel E and Clerc B. Running title: Albumin infusion in septic cirrhotic patients. Keywords: Liver Cirrhosis; Sepsis; Renal Failure; Portal Hypertension; Prognosis. Words: Abstract: 249. Abstract + Text + references + tables + figure legends: 5798 words. References: 37. Figures: 3. Tables: 4. Supplementary material for online publication: Appendix Files 1 to 4. Appendix Tables 1 to 3. Abbreviations: AKI, acute Kidney Injury; IQR, interquartile range; LVP, large-volume paracentesis;
Page 2, Thévenot T., et al. MDRD, modification of diet in renal disease; SBP, spontaneous bacterial peritonitis; SIRS, systemic inflammatory response syndrome. Competing Interests: We thank LFB laboratory (France) for partial financial support. Disclosures: No conflicts of interest exist. Funding: The French Hospital Clinical Research Program Clinical trial number: NCT01359813 Writing assistance: Fiona Ecarnot revised the English for payment. Address for correspondence: Thierry THEVENOT, Service d’Hépatologie et de Soins Intensifs Digestifs, University Hospital Jean Minjoz, 25000 Besançon, France. Tel: 33 3 81 66 84 21, FAX: 33 3 81 66 84 18, E-mail: [email protected]
Page 3, Thévenot T., et al.
Abstract Background&Aims: Albumin infusion improves renal function and survival in cirrhotic patients with spontaneous bacterial peritonitis (SBP) but its efficacy in other types of infections remains unknown. We investigated this issue through a multicenter randomized controlled trial. Methods: A total of 193 cirrhotic patients with a Child-Pugh score greater than 8 and sepsis unrelated to SBP were randomly assigned to receive antibiotics plus albumin (1.5 g/kg on day 1 and 1 g/kg on day 3; albumin group [ALB]:n=96) or antibiotics alone (control group [CG]:n=97). The primary endpoint was the 3-month renal failure rate (increase in creatinine ≥50% to reach a final value ≥133µmol/L). The secondary endpoint was 3-month survival rate. Results: Forty-seven (24.6%) patients died (ALB: n=27 vs CG: n=20; 3-month survival: 70.2% vs. 78.3%; p=0.16). Albumin infusion delayed the occurrence of renal failure (mean time to onset, ALB: 29.0±21.8 vs 11.7±9.1 days, p=0.018) but the 3month renal failure rate was similar (ALB: 14.3% vs CG: 13.5%; p=0.88). By multivariate analysis, MELD score (p160 µmol/L, SBP, septic shock, and use of antibiotics (except antibiotic prophylaxis) during the week prior randomization. We excluded patients with a creatinine level >160 µmol/L because
the Scientific
Committee considered that
randomization would be unethical and not feasible in these patients. Randomization Randomization was performed in blocks of 4, in a 1:1 ratio, by means of an interactive voicesystem response (Ascopharm, Novasco group, Paris, France) with stratification by center. The number of subjects per block was known only to the methodologist. Treatment allocation and follow-up Eligible patients were randomly assigned to receive either antibiotics alone (control group) or antibiotics plus albumin 20% (ALB group: 1.5 g/kg on day 1 and 1 g/kg on day 3). Albumin administration was initiated in the first 12 hours after randomization, and antibiotics were to be started as soon as possible without waiting the randomization. During the 3-month followup period, eight visits were planned (at days 1, 3, 6, 9, 15, 30, 60 and 90) to record clinical and biological data (Supplementary Table 1). Complications of cirrhosis were managed according to standard protocols [20]. In septic patients with ascites, diuretics had to be stopped at the time of infection and re-introduced after resolution of infection. Large-volume paracentesis (LVP > 3 liters) was not authorized until after resolution of infection, but paracentesis of lesser volumes without albumin administration was authorized in cases of abdominal discomfort. Proposals for the choice of antibiotics was given in the study protocol to assist physicians (Supplementary File 3).
Endpoints and definitions Endpoints The primary endpoint was the rate of renal failure during the 3-month follow-up period. The secondary end-point was 3-month mortality rate. Renal failure
Page 7, Thévenot T., et al. Renal function was assessed by measuring serum creatinine concentration at inclusion and throughout follow-up (see Supplementary Table 1 for measurement schedule). To define renal failure, a cut-off serum creatinine level of 133 µmol/L (15 mg/L) was used. For patients without pre-existing renal insufficiency, renal impairment was diagnosed whenever there was an increase in serum creatinine of ≥50%, reaching a final value of over 133 µmol/L. For patients with pre-existing renal insufficiency before infection, renal impairment was diagnosed by an increase in the serum creatinine level by more than 50% from baseline. Glomerular filtration rate was estimated according to the MDRD formula [21]. Statistical analysis Quantitative variables are presented as mean±SD or as median and interquartile range (IQR), and categorical variables as number (percentage). The sample size was estimated based on a projected 27% renal failure rate at 3 months in the control group [9] and 10% in the albumin infusion group [6]. It was calculated that 186 patients (93 per group) would yield 80% power with a two-sided alpha risk of 5%. The sample size was increased by 10% to 206 patients to allow for possible drop-outs, protocol deviations and patients lost to follow-up. All analyses were performed according to the intention-to-treat principle. Continuous variables were compared using the Student t-test or the Mann-Whitney test, as appropriate, and categorical variables using the Chi-squared test or Fisher’s Exact test as appropriate. For primary and secondary end-points, the Kaplan Meier method was used to estimate cumulative rates over the 3-month follow-up. For renal failure hazards, the time from date of inclusion to the date of the first increase in serum creatinine concentration defining renal failure was calculated, and occurrence of death, liver transplantation or study drop-out during follow-up were treated as censoring events. Curves for cumulative rates in both treatment arms were compared using the log-rank test. The same method was applied for the comparison of 3-month survival. Univariate analysis identified predictors of renal failure and death, which were then included in multivariate analysis using Cox’s logistic regression model. All statistical analyses were performed using
Page 8, Thévenot T., et al. SAS 9.0 (SAS Institute Inc., Cary, NC, USA). A p values 90 /min Respiratory rate > 20/min or PaCO2< 32 mmHg
0.45
0.69 0.52
Leukocytes < 4,000 or > 12,000/mm3 or immature neutrophils >10%
Biliary or GI tract infection
Nosocomial / community acquired infection Positive blood culture
0.65
Page 30, Thévenot T., et al.
Table 3. Baseline characteristics according to the occurrence of renal failure during the 3-month follow-up period No renal failure
renal failure
n =155
n = 25
Patients treated with albumin
78 (50.3)
13 (52.0)
0.87
Mean age (years)
54.7 ± 8.2
59.4 ± 10.1
0.011
Male gender
106 (68.4)
17 (68.0)
0.97
Etiology of cirrhosis: Alcohol vs. other
142 (91.6)
23 (92.0)
1.00
Child-Pugh score
11.1 ± 1.4
11.0 ± 1.3
0.66
MELD score
20.9 ± 5.7
21.9 ± 5.9
0.42
43.5 ± 13.3
41.8 ± 12.0
0.53
24.5 ± 5.0
23.2 ± 4.6
0.22
91 (51.0-161.0)
110 (50.0-194.5)
0.41
Ascites
109 (70.3)
14 (56.0)
0.15
Sodium (mmol/L)
132.9 ± 5.1
133.0 ± 7.1
0.93
Creatinine (µmol/L)
74.3 ± 30.0
83.2 ± 29.4
0.17
111.8 ± 50.7
94.7 ± 44.8
0.11
34 (16.9-58.8)
26.6 (15.3-63)
0.94
Severe acute alcoholic hepatitis
25 (16.1)
3 (12.0)
0.77
Pneumonia
52 (33.5)
8 (32.0)
0.88
Culture negative infection
58 (37.4)
18 (72.0)
0.0012
Severe sepsis
18 (11.6)
7 (28.0)
0.054
Prothrombin time Albumin (g/L) Total bilirubin (µmol/L)
#
Glomerular filtration rate (mL/min) C-reactive protein (mg/L)
#
p
Page 31, Thévenot T., et al.
Table 4 : Factors independently related to 3-month survival by multivariate analysis (Cox model)
HR§
95% CI
p
MELD score*
1.10
1.05- 1.16
S0168-8278(14)00862-9 http://dx.doi.org/10.1016/j.jhep.2014.11.017 JHEPAT 5438
To appear in:
Journal of Hepatology
Received Date: Revised Date: Accepted Date:
13 August 2014 5 November 2014 6 November 2014
Please cite this article as: Thévenot, T., Bureau, C., Oberti, F., Anty, R., Louvet, A., Plessier, A., Rudler, M., Heurgué-Berlot, A., Rosa, I., Talbodec, N., Dao, T., Ozenne, V., Carbonell, N., Causse, X., Goria, O., Minello, A., De Ledinghen, V., Amathieu, R., Barraud, H., Nguyen-Khac, E., Becker, C., Paupard, T., Botta-Fridlung, D., Abdelli, N., Guillemot, F., Monnet, E., Di Martino, V., Effect of Albumin in Cirrhotic Patients with Infection Other Than Spontaneous Bacterial Peritonitis. A Randomized Trial, Journal of Hepatology (2014), doi: http://dx.doi.org/ 10.1016/j.jhep.2014.11.017
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Page 1, Thévenot T., et al. EFFECT OF ALBUMIN IN CIRRHOTIC PATIENTS WITH INFECTION OTHER THAN SPONTANEOUS BACTERIAL PERITONITIS. A RANDOMIZED TRIAL. Thierry Thévenot (1), Christophe Bureau (2), Frédéric Oberti (3), Rodolphe Anty (4), Alexandre Louvet (5), Aurélie Plessier (6), Marika Rudler (7), Alexandra Heurgué-Berlot (8),Isabelle Rosa (9), Nathalie Talbodec (10), Thong Dao (11), Violaine Ozenne (12), Nicolas Carbonell (13), Xavier Causse (14), Odile Goria (15), Anne Minello (16), Victor De Ledinghen (17), Roland Amathieu (18), Hélène Barraud(19), Eric Nguyen-Khac (20), Claire Becker (21), Thierry Paupard (22), Danielle Botta-Fridlung (23), Naceur Abdelli (24), François Guillemot (25), Elisabeth Monnet (1), Vincent Di Martino (1). d'Hépatologie, hôpital Jean Minjoz, 25000 Besançon, (2)Service d’HépatoGastroentérologie, hôpital Purpan, clinique Dieulafoy, 31059 Toulouse, (3)Service d’ d’HépatoGastroentérologie, hôpital d’Angers, 4 rue Larrey 49100 Angers, (4)Service d’HépatoGastroentérologie, hôpital Archet, rue St Antoine Ginestier, 06200 Nice, (5)Service des Maladies de l’Appareil Digestif, CHRU de Lille, rue M. Polonovski, 59037 Lille cedex (6)Inserm U-773, Service d’Hépatologie, hôpital Beaujon, 100 boulevard du Général Leclerc, 92118 Clichy cédex, (7)Service d’Hépato-Gastroentérologie, hôpital Pitié-Salpétrière, 47-83 boulevard de l’Hôpital, 75651 Paris Cedex 13, (8)Service d’Hépato-Gastroentérologie, hôpital Robert Debré, 51092 Reims cedex, (9)Service d’Hépato-Gastroentérologie, CHIC de Créteil, 40 avenue de Verdun, 94010 Créteil cedex, (10)Service d’Hépato-Gastroentérologie, centre hospitalier, 135 rue du Président Coty, 59 200 Tourcoing, (11)Inserm U-1075, Service d’Hépato-Gastroentérologie, hôpital de Caen, avenue de la Côte de Nacre, 14000 Caen, (12)Service d’Hépato-Gastroentérologie, hôpital Lariboisière, 2 rue AmbroiseParé, 75010 Paris, (13)Service d’Hépato-Gastroentérologie, hôpital Saint-Antoine, 184 rue du Fbg St Antoine, 75571 Paris cedex 12, (14)Service d’Hépato-Gastroentérologie, hôpital de la Source, BP 6709, 45067 Orléans cedex 12, (15)Service d’Hépato-Gastroentérologie, hôpital de Rouen, 1 Rue Germont, 76000 Rouen, (16)Service d’Hépato-Gastroentérologie, hôpital du Bocage, BP 1542, 21034 Dijon cedex, (17)Service d’Hépato-Gastroentérologie, hôpital du haut Levêque, 33604 Pessac Cedex, (18)Service d’Hépato-Gastroentérologie, hôpital Jean Verdier, av. du 14 juillet, 93143 Bondy cedex, (19)Service d’Hépato-Gastroentérologie, hôpital de Brabois, rue du Morvan, 54511 Vandoeuvre-lesNancy cedex, (20)Service d’Hépato-Gastroentérologie, hôpital d’Amiens, 2 Place Victor Pauchet, 80080 Amiens, (21)Service d’Hépato-Gastroentérologie, hôpital de Lens, 99 route de la Bassée, SP-8, 62307 Lens cedex, (22)Service d’Hépato-Gastroentérologie, hôpital de Dunkerque, 130 avenue Louis Herbeaux, 59385 Dunkerque cedex 1, (23)Service d’Hépato-Gastroentérologie, hôpital de la Conception, 147 Bd Baille, 13005 Marseille, (24)Service d’Hépato-Gastroentérologie, hôpital de Chalons-en-Champagne, 51 rue du commandant Derrien, 51005 Chalons-en-Champagne cedex, (25)Service d’Hépato-Gastroentérologie, hôpital de Roubaix, 11-17 Boulevard Lacordaire, 59100 Roubaix. ( 1)Service
Study concept and design : Thevenot T, Di Martino V and Monnet E; Acquisition of data : Thévenot T, Bureau C, Oberti F, Anty R, Louvet A, Plessier A, Rudler M, Heurgué-Berlot A, Rosa I, Talbodec N, Dao T, Ozenne V, Carbonell N, Causse X, Goria O, Minello A, De Ledinghen V, Amathieu R, Barraud H, Nguyen-Khac E, Becker C, Paupard T, Botta-Fridlung D, Abdelli N, Guillemot F; Analysis, interpretation of data and statistical analysis : Thevenot T and Monnet E; drafting of the manuscript and obtained funding : Thevenot T; Technical support: Leroux F, Muel E and Clerc B. Running title: Albumin infusion in septic cirrhotic patients. Keywords: Liver Cirrhosis; Sepsis; Renal Failure; Portal Hypertension; Prognosis. Words: Abstract: 249. Abstract + Text + references + tables + figure legends: 5798 words. References: 37. Figures: 3. Tables: 4. Supplementary material for online publication: Appendix Files 1 to 4. Appendix Tables 1 to 3. Abbreviations: AKI, acute Kidney Injury; IQR, interquartile range; LVP, large-volume paracentesis;
Page 2, Thévenot T., et al. MDRD, modification of diet in renal disease; SBP, spontaneous bacterial peritonitis; SIRS, systemic inflammatory response syndrome. Competing Interests: We thank LFB laboratory (France) for partial financial support. Disclosures: No conflicts of interest exist. Funding: The French Hospital Clinical Research Program Clinical trial number: NCT01359813 Writing assistance: Fiona Ecarnot revised the English for payment. Address for correspondence: Thierry THEVENOT, Service d’Hépatologie et de Soins Intensifs Digestifs, University Hospital Jean Minjoz, 25000 Besançon, France. Tel: 33 3 81 66 84 21, FAX: 33 3 81 66 84 18, E-mail: [email protected]
Page 3, Thévenot T., et al.
Abstract Background&Aims: Albumin infusion improves renal function and survival in cirrhotic patients with spontaneous bacterial peritonitis (SBP) but its efficacy in other types of infections remains unknown. We investigated this issue through a multicenter randomized controlled trial. Methods: A total of 193 cirrhotic patients with a Child-Pugh score greater than 8 and sepsis unrelated to SBP were randomly assigned to receive antibiotics plus albumin (1.5 g/kg on day 1 and 1 g/kg on day 3; albumin group [ALB]:n=96) or antibiotics alone (control group [CG]:n=97). The primary endpoint was the 3-month renal failure rate (increase in creatinine ≥50% to reach a final value ≥133µmol/L). The secondary endpoint was 3-month survival rate. Results: Forty-seven (24.6%) patients died (ALB: n=27 vs CG: n=20; 3-month survival: 70.2% vs. 78.3%; p=0.16). Albumin infusion delayed the occurrence of renal failure (mean time to onset, ALB: 29.0±21.8 vs 11.7±9.1 days, p=0.018) but the 3month renal failure rate was similar (ALB: 14.3% vs CG: 13.5%; p=0.88). By multivariate analysis, MELD score (p160 µmol/L, SBP, septic shock, and use of antibiotics (except antibiotic prophylaxis) during the week prior randomization. We excluded patients with a creatinine level >160 µmol/L because
the Scientific
Committee considered that
randomization would be unethical and not feasible in these patients. Randomization Randomization was performed in blocks of 4, in a 1:1 ratio, by means of an interactive voicesystem response (Ascopharm, Novasco group, Paris, France) with stratification by center. The number of subjects per block was known only to the methodologist. Treatment allocation and follow-up Eligible patients were randomly assigned to receive either antibiotics alone (control group) or antibiotics plus albumin 20% (ALB group: 1.5 g/kg on day 1 and 1 g/kg on day 3). Albumin administration was initiated in the first 12 hours after randomization, and antibiotics were to be started as soon as possible without waiting the randomization. During the 3-month followup period, eight visits were planned (at days 1, 3, 6, 9, 15, 30, 60 and 90) to record clinical and biological data (Supplementary Table 1). Complications of cirrhosis were managed according to standard protocols [20]. In septic patients with ascites, diuretics had to be stopped at the time of infection and re-introduced after resolution of infection. Large-volume paracentesis (LVP > 3 liters) was not authorized until after resolution of infection, but paracentesis of lesser volumes without albumin administration was authorized in cases of abdominal discomfort. Proposals for the choice of antibiotics was given in the study protocol to assist physicians (Supplementary File 3).
Endpoints and definitions Endpoints The primary endpoint was the rate of renal failure during the 3-month follow-up period. The secondary end-point was 3-month mortality rate. Renal failure
Page 7, Thévenot T., et al. Renal function was assessed by measuring serum creatinine concentration at inclusion and throughout follow-up (see Supplementary Table 1 for measurement schedule). To define renal failure, a cut-off serum creatinine level of 133 µmol/L (15 mg/L) was used. For patients without pre-existing renal insufficiency, renal impairment was diagnosed whenever there was an increase in serum creatinine of ≥50%, reaching a final value of over 133 µmol/L. For patients with pre-existing renal insufficiency before infection, renal impairment was diagnosed by an increase in the serum creatinine level by more than 50% from baseline. Glomerular filtration rate was estimated according to the MDRD formula [21]. Statistical analysis Quantitative variables are presented as mean±SD or as median and interquartile range (IQR), and categorical variables as number (percentage). The sample size was estimated based on a projected 27% renal failure rate at 3 months in the control group [9] and 10% in the albumin infusion group [6]. It was calculated that 186 patients (93 per group) would yield 80% power with a two-sided alpha risk of 5%. The sample size was increased by 10% to 206 patients to allow for possible drop-outs, protocol deviations and patients lost to follow-up. All analyses were performed according to the intention-to-treat principle. Continuous variables were compared using the Student t-test or the Mann-Whitney test, as appropriate, and categorical variables using the Chi-squared test or Fisher’s Exact test as appropriate. For primary and secondary end-points, the Kaplan Meier method was used to estimate cumulative rates over the 3-month follow-up. For renal failure hazards, the time from date of inclusion to the date of the first increase in serum creatinine concentration defining renal failure was calculated, and occurrence of death, liver transplantation or study drop-out during follow-up were treated as censoring events. Curves for cumulative rates in both treatment arms were compared using the log-rank test. The same method was applied for the comparison of 3-month survival. Univariate analysis identified predictors of renal failure and death, which were then included in multivariate analysis using Cox’s logistic regression model. All statistical analyses were performed using
Page 8, Thévenot T., et al. SAS 9.0 (SAS Institute Inc., Cary, NC, USA). A p values 90 /min Respiratory rate > 20/min or PaCO2< 32 mmHg
0.45
0.69 0.52
Leukocytes < 4,000 or > 12,000/mm3 or immature neutrophils >10%
Biliary or GI tract infection
Nosocomial / community acquired infection Positive blood culture
0.65
Page 30, Thévenot T., et al.
Table 3. Baseline characteristics according to the occurrence of renal failure during the 3-month follow-up period No renal failure
renal failure
n =155
n = 25
Patients treated with albumin
78 (50.3)
13 (52.0)
0.87
Mean age (years)
54.7 ± 8.2
59.4 ± 10.1
0.011
Male gender
106 (68.4)
17 (68.0)
0.97
Etiology of cirrhosis: Alcohol vs. other
142 (91.6)
23 (92.0)
1.00
Child-Pugh score
11.1 ± 1.4
11.0 ± 1.3
0.66
MELD score
20.9 ± 5.7
21.9 ± 5.9
0.42
43.5 ± 13.3
41.8 ± 12.0
0.53
24.5 ± 5.0
23.2 ± 4.6
0.22
91 (51.0-161.0)
110 (50.0-194.5)
0.41
Ascites
109 (70.3)
14 (56.0)
0.15
Sodium (mmol/L)
132.9 ± 5.1
133.0 ± 7.1
0.93
Creatinine (µmol/L)
74.3 ± 30.0
83.2 ± 29.4
0.17
111.8 ± 50.7
94.7 ± 44.8
0.11
34 (16.9-58.8)
26.6 (15.3-63)
0.94
Severe acute alcoholic hepatitis
25 (16.1)
3 (12.0)
0.77
Pneumonia
52 (33.5)
8 (32.0)
0.88
Culture negative infection
58 (37.4)
18 (72.0)
0.0012
Severe sepsis
18 (11.6)
7 (28.0)
0.054
Prothrombin time Albumin (g/L) Total bilirubin (µmol/L)
#
Glomerular filtration rate (mL/min) C-reactive protein (mg/L)
#
p
Page 31, Thévenot T., et al.
Table 4 : Factors independently related to 3-month survival by multivariate analysis (Cox model)
HR§
95% CI
p
MELD score*
1.10
1.05- 1.16
