Publication

• Title: Effect of Intravenous Albumin on Renal Impairment and Mortality in Patients with Cirrhosis and Spontaneous Bacterial Peritonitis
• Acronym: None
• Year: 1999
• Journal published in: The New England Journal of Medicine
• Citation: Sort P, Navasa M, Arroyo V, et al. Effect of intravenous albumin on renal impairment and mortality in patients with cirrhosis and spontaneous bacterial peritonitis. N Engl J Med. 1999;341(6):403-9.

Context & Rationale

• Background

  Spontaneous bacterial peritonitis (SBP) is a frequent, high-mortality complication of decompensated cirrhosis with ascites.
  Even with effective non-nephrotoxic antibiotics, approximately one-third of patients develop renal impairment, which is the strongest short-term prognostic marker.
  Contemporary mechanistic framing (at the time) linked SBP-associated renal dysfunction to further deterioration of “effective arterial blood volume” and amplification of neurohumoral vasoconstrictor pathways (e.g., renin–angiotensin activation) in an already vasodilated cirrhotic circulation.
  Albumin was already established as an effective plasma expander in other cirrhosis contexts (e.g., post-paracentesis circulatory dysfunction), but its role as adjunctive therapy in SBP to prevent renal impairment and improve survival was uncertain.
• Research Question/Hypothesis

  In adults with cirrhosis and SBP, does early intravenous albumin, when added to standard antibiotic therapy, reduce (i) renal impairment during hospitalisation and (ii) mortality, compared with antibiotic therapy alone?
• Why This Matters

  SBP-related renal failure (functionally overlapping with hepatorenal syndrome physiology) is a key “final common pathway” to death in advanced cirrhosis.
  Demonstrating an effective, deliverable adjunct that prevents renal failure and improves survival would redefine SBP as not solely an infectious disease problem, but a circulatory/organ-failure syndrome requiring targeted haemodynamic support.

Design & Methods

• Research Question: Whether early plasma volume expansion with intravenous albumin, added to cefotaxime, prevents renal impairment and reduces mortality in cirrhosis with SBP.
• Study Type: Multicentre, randomised, parallel-group, open-label controlled trial; investigator-initiated; 7 university hospitals (Spain); enrolment Nov 1995–Sep 1997; central review committee verified analyses.
• Population:
  • Setting: Hospitalised adults with cirrhosis and ascites diagnosed with SBP.
  • Key inclusion: Ascitic-fluid polymorphonuclear count >250/mm³ (with bedside inoculation into blood-culture bottles); age 18–80 years.
  • Key exclusions: Antibiotics within 1 week pre-diagnosis (except prophylactic norfloxacin); evidence of secondary peritonitis; other infections; shock; gastrointestinal bleeding; ileus; grade 3–4 encephalopathy; cardiac failure; findings suggestive of organic nephropathy (proteinuria/haematuria/abnormal renal ultrasound); HIV; advanced neoplasia affecting short-term prognosis; serum creatinine >3 mg/dl.
  • Volume status safeguard: If dehydration suspected, central venous pressure was measured; patients with CVP <4 mm Hg were excluded.
  • Numbers: 199 screened; 73 excluded; 126 randomised (63 per group).
• Intervention:
  • Antibiotic: Intravenous cefotaxime, dose-adjusted to creatinine (2 g every 6 h if creatinine <1.5 mg/dl; 1 g every 6 h if 1.5–2.0 mg/dl; 1 g every 8 h if >2.0–2.5 mg/dl; 1 g every 12 h if >2.5 mg/dl).
  • Albumin: 20% human albumin, 1.5 g/kg within the first 6 h after enrolment, then 1 g/kg on day 3 (day-3 labs drawn before albumin administration).
  • SBP resolution definition: Disappearance of infection signs and ascitic PMN ≤250/mm³.
• Comparison:
  • Identical cefotaxime regimen without albumin infusion at SBP diagnosis.
  • Antibiotic modification permitted in non-responders (culture-guided if organism isolated; empirical if cultures negative).
  • Diuretics and therapeutic paracentesis prohibited until infection resolution; partial paracentesis (3 L) occurred pre-resolution in 7 patients with tense ascites (3 control; 4 albumin arm).
  • After infection resolution: prophylactic norfloxacin 400 mg/day continued throughout follow-up; tense ascites treated with total paracentesis plus albumin (6–8 g/L removed) irrespective of original allocation.
• Blinding: Unblinded (open-label). Randomisation performed using sealed envelopes; central review committee verified analyses; primary endpoints were largely objective (biochemistry and mortality), but care delivery decisions were potentially influenceable by lack of blinding.
• Statistics: Power calculation: assuming renal impairment in ~30% with cefotaxime, 50 patients/group required to detect a 25% difference between groups in renal impairment during hospitalisation with two-sided α=0.05 and β=0.20 (80% power); final analysis intention-to-treat; univariate testing (χ²/Fisher/t-test) plus stepwise logistic regression for independent predictors.
• Follow-Up Period: In-hospital follow-up; post-discharge weekly for 1 month then monthly to 90 days from enrolment.

Key Results

This trial was not stopped early. Recruitment and follow-up proceeded to the planned completion (126 randomised).

• Outcome divergence occurred despite similar infection resolution: 94% vs 98% resolution (P=0.36), but renal impairment 33% vs 10% (P=0.002) and in-hospital death 29% vs 10% (P=0.01).
• Renal impairment was a dominant mediator of early mortality: in-hospital death 21/27 (78%) with renal impairment vs 3/99 (3%) without (P<0.001); 3-month death 24/27 (89%) vs 16/99 (16%) (P<0.001).
• Risk-stratified signal (as reported): renal impairment among those with baseline bilirubin ≥4 mg/dl: 14/29 (48%) vs 3/25 (12%); among bilirubin <4 mg/dl with creatinine ≥1 mg/dl: 6/19 (32%) vs 3/21 (14%); among bilirubin <4 mg/dl with creatinine <1 mg/dl: 7% vs 0%.

Internal Validity

• Randomisation and allocation concealment: randomisation list generated by SAS; sealed envelopes at each hospital; investigators unaware of assignments at the point of randomisation; open-label thereafter (risk of performance bias remains).
• Screening and selection: 199 assessed; 73 excluded on clinical/eligibility grounds (including shock, bleeding, severe encephalopathy, creatinine >3 mg/dl, suspected dehydration with CVP <4 mm Hg), yielding a physiologically “stabilised” SBP cohort; this strengthens internal coherence but narrows case-mix.
• Post-randomisation exclusions/withdrawals: 2 patients in the albumin arm did not meet inclusion criteria and were withdrawn from that arm’s protocolised albumin dosing; trial reports intention-to-treat analysis for the final dataset of 126.
• Blinding and bias: unblinded design could influence co-interventions (e.g., timing of paracentesis, escalation of supportive care), yet key endpoints were objective (renal biochemistry and mortality).
• Protocol adherence and contamination: after infection resolution, both groups could receive albumin with total paracentesis (6–8 g/L removed); this is a pre-specified “post-SBP” management pathway and would tend to reduce between-group separation rather than inflate it.
• Baseline comparability: no significant differences reported across baseline characteristics; renal failure at enrolment 28/63 (44%) vs 25/63 (40%); Child–Pugh score 10±0.2 vs 10±0.2.
• Separation of the variable of interest (physiological separation): BUN (mg/dl) day 3: 34±3 vs 25±3 (P=0.03); day 6: 36±3 vs 22±3 (P=0.003); day 9: 36±3 vs 22±3 (P=0.01).
• Separation of the variable of interest (renal function): creatinine (mg/dl) day 6: 1.3±0.1 vs 1.0±0.1 (P=0.03); day 9: 1.4±0.1 vs 1.0±0.1 (P=0.04).
• Separation of the variable of interest (serum sodium): sodium (mmol/l) day 3: 130±1 vs 134±1 (P=0.001); day 6: 130±1 vs 134±1 (P<0.001); day 9: 130±1 vs 134±1 (P=0.002); mean arterial pressure did not differ across days 0–9.
• Outcome definitions: renal impairment required “nonreversible” deterioration during hospitalisation with thresholded >50% rises; this is clinically meaningful but differs from contemporary acute kidney injury staging systems (potential interpretive drift when mapping to modern practice).
• Missing data/follow-up: 7 patients lost after discharge (4 vs 3); 90-day mortality calculated as known deaths divided by total randomised, which assumes non-ascertainment is non-informative and may bias absolute 90-day mortality estimates.
• Statistical rigor: ITT stated; multivariable modelling used to identify independent predictors; effect estimates for key endpoints provided as adjusted odds ratios for renal impairment and in-hospital mortality.

Conclusion on Internal Validity: Moderate-to-strong internal validity: randomisation and objective endpoints support causal inference, and there was clear physiological and clinical separation; limitations mainly arise from the open-label design, substantial pre-randomisation exclusions, and a renal outcome definition that does not map one-to-one to contemporary AKI criteria.

External Validity

• Population representativeness: hospitalised cirrhosis with SBP in Spanish university centres; high baseline severity (Child–Pugh ~10; renal failure at enrolment ~40%); excluded patients with shock, active GI bleeding, advanced encephalopathy, marked renal dysfunction (creatinine >3 mg/dl), and suspected hypovolaemia (CVP <4 mm Hg).
• Practice evolution: cefotaxime-based therapy was standard at the time; current antimicrobial choices may differ due to changing resistance patterns and health-care associated SBP epidemiology, but the pathophysiological target (circulatory dysfunction precipitating renal failure) remains relevant.
• Intervention feasibility: albumin dosing is operationally straightforward (weight-based, day 1 and day 3), but resource availability and cost considerations may influence adoption in low-resource settings.
• Applicability across SBP phenotypes: findings most directly apply to haemodynamically stable SBP without major competing acute insults (shock/bleeding), and may be less directly generalisable to ICU-level septic shock SBP or multifactorial AKI.

Conclusion on External Validity: Generalisability is good for “classical” inpatient SBP cohorts without shock or advanced non-cirrhotic renal disease, but extrapolation to modern nosocomial SBP with multidrug resistance, and to severely unstable presentations, requires clinical judgement.

Strengths & Limitations

• Strengths: clinically decisive endpoints (renal impairment and mortality); pragmatic dosing regimen; multicentre recruitment; intention-to-treat analysis; mechanistic alignment (renin activity and renal outcomes); clear between-group separation in renal physiology and outcomes.
• Limitations: unblinded design; moderate sample size with wide confidence intervals for adjusted effects; restrictive exclusions (potentially selecting a “cleaner” haemodynamic phenotype); renal impairment definition is era-specific; 90-day mortality ascertainment incomplete for 7 patients with a denominator-based approach that may dilute event rates.

Interpretation & Why It Matters

• Causal framing

  The trial supports a model in which SBP mortality is not explained solely by infection control; prevention of circulatory deterioration and renal failure is a critical therapeutic target, as shown by large outcome differences despite near-identical infection resolution.
• Clinical practice impact

  Albumin became a core adjunct to antibiotics in SBP management, particularly for patients at appreciable risk of renal impairment (e.g., hyperbilirubinaemia and/or baseline renal dysfunction), shifting “standard care” from antimicrobial monotherapy to combined infection + circulatory support.
• Mechanistic coherence

  Lower BUN/creatinine trajectories and higher sodium in the albumin group over days 3–9, alongside lower renin activation, align with effective plasma volume expansion mitigating neurohumoral vasoconstrictor escalation, even without detectable mean arterial pressure differences.

Controversies & Subsequent Evidence

• “Who needs albumin?” (risk targeting vs universal use): the accompanying editorial highlighted that the most plausible clinical strategy is to prioritise albumin for higher-risk SBP presentations given the clear risk stratification by baseline biochemistry and the non-trivial dose/cost burden of 1.5 g/kg then 1 g/kg albumin.¹
• Cost-effectiveness and low-risk event rates: correspondence argued that low-risk subsets (e.g., lower bilirubin and preserved renal function) have very low renal impairment rates, raising the possibility of overtreatment if albumin is applied indiscriminately, and reinforcing the importance of baseline risk assessment.²
• Open-label design and co-interventions: the lack of blinding was an inherent vulnerability; however, the magnitude of effect on objective endpoints and the mechanistic consistency (renal indices and neurohumoral activation) make pure performance-bias explanations less persuasive.
• Subsequent evidence landscape (high-level): later syntheses and practice guidance documents in hepatology broadly incorporated the concept that albumin plus antibiotics reduces SBP-associated renal dysfunction and mortality, and commonly recommend the same weight-based dosing schedule tested here (details: see Further Reading).

Summary

• In 126 hospitalised adults with cirrhosis and SBP, adding intravenous albumin (1.5 g/kg within 6 h + 1 g/kg on day 3) to cefotaxime reduced renal impairment from 33% to 10% (P=0.002).
• In-hospital mortality fell from 29% to 10% (P=0.01), and 3-month mortality from 41% to 22% (P=0.03), with infection resolution similar between groups (94% vs 98%).
• Renal impairment was the major determinant of short-term death: 78% in-hospital mortality with renal impairment vs 3% without (P<0.001).
• Between-group physiological separation was evident (days 3–9): lower BUN/creatinine and higher sodium in the albumin arm, without differences in mean arterial pressure.
• Main interpretive constraints are open-label design, selection via exclusions, and an era-specific renal impairment definition; nevertheless, objective endpoints and effect magnitude support a clinically meaningful treatment effect.

Further Reading

Other Trials

• 1990Ginès P, Titó L, Arroyo V, et al. Norfloxacin prevents spontaneous bacterial peritonitis recurrence in cirrhosis: results of a double-blind placebo-controlled trial. Hepatology. 1990;12:716-724.
• 1991Runyon BA. Short-course versus long-course antibiotic treatment of spontaneous bacterial peritonitis: a randomized controlled study of 100 patients. Gastroenterology. 1991;100:1737-1742.
• 1985Felisart J, Rimola A, Arroyo V, et al. Cefotaxime is more effective than is ampicillin-tobramycin in cirrhosis with severe infections. Hepatology. 1985;5:457-462.
• 1995Rimola A, Salmerón JM, Clemente G, et al. Two different dosages of cefotaxime in the treatment of spontaneous bacterial peritonitis in cirrhosis: results of a prospective, randomized, multicenter study. Hepatology. 1995;21:674-679.
• 1996Navasa M, Fernández J, Rodés J. Randomized comparative study of oral ofloxacin versus intravenous cefotaxime in spontaneous bacterial peritonitis. Gastroenterology. 1996;111:1011-1017.

Systematic Review & Meta Analysis

• 2013Salerno F, Navickis RJ, Wilkes MM. Albumin infusion improves outcomes of patients with spontaneous bacterial peritonitis: a meta-analysis of randomized trials. Clin Gastroenterol Hepatol. 2013;11:123-130.e1.
• 2012Role of albumin treatment in patients with spontaneous bacterial peritonitis. Clin Gastroenterol Hepatol. 2012. Not reported (bibliographic details not available in provided source materials).
• 2022Use of human albumin infusion in cirrhotic patients with spontaneous bacterial peritonitis: a systematic review and meta-analysis. Dig Dis Sci. 2022. Not reported (bibliographic details not available in provided source materials).
• 2011Tandon P, Garcia-Tsao G. Renal dysfunction is the most important independent predictor of mortality in cirrhotic patients with spontaneous bacterial peritonitis. Clin Gastroenterol Hepatol. 2011;9:260-265.
• 1999Bass NM. Intravenous albumin for spontaneous bacterial peritonitis in patients with cirrhosis. N Engl J Med. 1999;341:443-444.

Observational Studies

• 1994Follo A, Llovet JM, Navasa M, et al. Renal impairment after spontaneous bacterial peritonitis in cirrhosis: incidence, clinical course, predictive factors, and prognosis. Hepatology. 1994;20:1495-1501.
• 1998Navasa M, Follo A, Filella X, et al. Tumor necrosis factor and interleukin-6 in spontaneous bacterial peritonitis: relationship with the development of renal impairment and mortality. Hepatology. 1998;27:1227-1232.
• 1988Llach J, Ginès P, Arroyo V, et al. Prognostic value of arterial pressure, endogenous vasoconstrictor systems, and renal function in cirrhotic patients admitted to the hospital for the treatment of ascites. Gastroenterology. 1988;94:482-487.
• 2011Tandon P, Garcia-Tsao G. Renal dysfunction is the most important independent predictor of mortality in cirrhotic patients with spontaneous bacterial peritonitis. Clin Gastroenterol Hepatol. 2011;9:260-265.
• 1990Ginès P, Titó L, Arroyo V, et al. Norfloxacin prevents spontaneous bacterial peritonitis recurrence in cirrhosis: results of a double-blind placebo-controlled trial. Hepatology. 1990;12:716-724.

Guidelines

• 2021AASLD Practice Guidance. Diagnosis, evaluation, and management of ascites, spontaneous bacterial peritonitis and hepatorenal syndrome. Hepatology. 2021. Not reported (bibliographic details not available in provided source materials).
• 2010European Association for the Study of the Liver. EASL Clinical Practice Guidelines on the management of ascites, spontaneous bacterial peritonitis, and hepatorenal syndrome in cirrhosis. J Hepatol. 2010;53:397-417.
• 2018European Association for the Study of the Liver. EASL Clinical Practice Guidelines: management of patients with decompensated cirrhosis. J Hepatol. 2018;69:406-460.
• 1996Arroyo V, Ginès P, Gerbes AL, et al. Definition and diagnostic criteria of refractory ascites and hepatorenal syndrome in cirrhosis. Hepatology. 1996;23:164-176.
• 1999Patch D, Burroughs A. Intravenous albumin in patients with cirrhosis and spontaneous bacterial peritonitis. N Engl J Med. 1999;341:1773-1774.

Overall Takeaway

This trial reframed SBP from an infection-only problem to a syndrome in which preventing circulatory collapse and renal failure is central to survival. By showing large reductions in renal impairment and early mortality with a pragmatic, weight-based albumin regimen added to antibiotics, it established adjunctive albumin as a defining component of modern SBP management and influenced subsequent guideline-based care.

Bibliography

• 1.Bass NM. Intravenous albumin for spontaneous bacterial peritonitis in patients with cirrhosis. N Engl J Med. 1999;341(6):443-444.
• 2.Patch D, Burroughs A. Intravenous albumin in patients with cirrhosis and spontaneous bacterial peritonitis. N Engl J Med. 1999;341(23):1773-1774.