Choudhury
A Randomized Trial Comparing Terlipressin and Noradrenaline in Patients With Cirrhosis and Septic Shock
Choudhury. Liver International 2016, published on-line doi:10.1111/liv.13252
Clinical Question
- In patients with cirrhosis and septic shock does terlipressin in comparison with noradrenaline improve haemodynamic parameters?
Design
- Randomised controlled trial
- Randomised in 1:1 ratio
- Block randomisation with fixed block size of 6
- Sequentially numbered opaque sealed envelopes – not documented if any checks were completed to ensure allocation concealment was maintained
- Non-blinded
- Consecutive recruitment
- Non-inferiority trial
- Sample size calculation: 82 patients required to determine a non-inferiority limit of 15% from a baseline of 40% haemodynamic improvement in the noradrenaline group, with a two sided alpha of 0.05 and a power of 90%
- Intention to treat analysis used for primary outcome
- A per-protocol analysis was also performed
Setting
- Single liver intensive care unit, India
- Data collected March 2013 – September 2014
Population
- Inclusion criteria: Decompensated cirrhosis with septic shock, that did not respond to fluids
- Decompensated cirrhosis – not defined
- Septic shock
- 2+ SIRS criteria
- Proven/suspected infection
- Vasopressor required to maintain MAP ≥65mmHg despite 2 hours of fluid resuscitation
- 15ml/kg normal saline over 30min followed by 5% albumin at 50ml/hr for 6 hours.
- Exclusion criteria: age <18 years, valvular heart disease, coronary artery disease, acute mesenteric ischaemia, peripheral vascular disease, pregnancy, chronic kidney disease, uncontrolled or refractory variceal bleed, immunosuppressive drugs
- 511 patients screened of whom 84 randomised
- Comparing baseline characteristics of intervention vs. control group
- Age: 47 vs. 48
- Alcoholic liver disease: 64% vs. 52%
- Source of sepsis – significant difference
- Spontaneous bacterial peritonitis: 50% vs. 26.2%, p=0.04
- Pneumonia: 21.4% vs. 47.6%, p=0.02
- MELD score: 32 vs. 32
- Lactate: 3 vs. 3
Intervention
- Terlipressin
- Titrated and infused at a rate of 1.3-5.2ug/min (2-8mg over 24)
Control
- Noradrenaline
- Titrated and infused at a rate of 7.5-60ug/min
For all patients
- Standard medical therapy included IV fluids and antibiotics
- 20% albumin given at 1.5gm/kg on 1st day followed by 1mg/kg for next 2 days, except those anuric or on renal replacement therapy
- Target MAP ≥65mmHg
- If unable achieve MAP at highest dose of study drug, salvage therapy initiated
- Combination of noradrenaline at 7.5-30μg/min and terlipressin at 1.3μg-2.6ug/min.
- Hydrocortisone 50mg every 6 hours
- Adrenaline or phenylephrine used as per clinical need
- Slow low-efficient dialysis was done in cases with acidosis, worsening azotemia or anuria. Continuous renal replacement therapy not available during study period
Outcome
- Primary outcome:
- Achieve and maintain MAP >65 for initial 48 hours – significantly greater in terlipressin group
- 92.9% vs. 69.1%, p=0.005
- Number needed to treat: 5
- Fragility index: 5
- Achieve and maintain MAP >65 for initial 48 hours – significantly greater in terlipressin group
- Secondary outcomes – comparing terlipressin vs. noradrenaline groups
- Survival
- 48 hours – significantly higher in terlipressin group
- 95.2% vs. 71.4%, p=0.003
- 28 days – no significant difference
- 26.2% vs. 14.3%, p=0.17
- 48 hours – significantly higher in terlipressin group
- Target MAP achieved at 6 hours – no significant difference
- 90.5% vs. 78.6%, p=0.003
- Shock reversal at 48 hours (able to stop vasopressors) – significantly higher in terlipressin group
- 33.1% vs. 11.9%, p<0.02
- Need for salvage therapy – significantly higher in terlipressin group
- 54.7% vs. 33.3%, p=0.03
- Adverse events
- Overall – no significant difference
- 40.5% vs. 21.4%, p=0.06
- Peripheral cyanosis – significantly higher in terlipressin group
- 29% vs. 0%, p=0.05
- Overall – no significant difference
- Assessment of microcirculation during first 48 hours – trend towards improvement with terlipressin
- Central venous oxygen saturation, p=0.07
- VBG-ABG pCO2 difference, p=0.07
- Lactate clearance, p=0.08
- Length of ICU stay – no significant difference
- 6 vs. 5 days, p=0.99
- New onset variceal bleeding – significantly lower in terlipressin group
- 0% vs. 9.5%, p=0.01
- Per-protocol analysis (patients who could be maintained on monotherapy till the outcome)
- Success of therapy at 48 hours – no significant difference
- 100% vs. 82%, p=0.07
- 48 hour survival – no significant difference
- 100% vs. 89%, p=0.26
- 28 day survival – significantly higher in terlipressin group
- 47% vs. 11%, p=0.002
- Success of therapy at 48 hours – no significant difference
- Survival
Authors’ Conclusions
- Terlipressin is as effective as noradrenaline as a vasopressor in patients with cirrhosis and septic shock and has an early survival benefit
Strengths
- Randomised controlled trial
- Intention to treat and per-protocol analysis performed
- Study protocol published on clinicaltrials.gov
Weaknesses
- Single centre
- Non-blinded
- Number of baseline differences
- Large number of patients excluded limiting external validity
- Primary outcome was not a patient orientated outcome
- Dose of albumin recorded in gm/kg on day 1 and mg/kg on day 2. This is likely to be a typographical error
The Bottom Line
- In patients with cirrhosis and septic shock, this single centre non-blinded RCT demonstrated that terlipressin in comparison with noradrenaline improved haemodynamics and had a mortality benefit. Due to a number of methodological flaws and baseline differences I would want further evidence before this becomes standard practice.
External Links
- [article] A randomized trial comparing terlipressin and noradrenaline in patients with cirrhosis and septic shock
- [further reading] Noradrenaline vs. terlipressin in the treatment of hepatorenal syndrome: A randomized study
- [further reading] Terlipressin versus noradrenaline in the treatment of hepatorenal syndrome: systematic review with meta-analysis and full economic evaluation
- [further reading] Vasopressin versus Norepinephrine Infusion in Patients with Septic Shock
Metadata
Summary author: @davidslessor
Summary date: 5th March 2017
Peer-review editor: @avkwong
Pingback: Liu – The Bottom Line