hepatorenal syndrome

admission bundle

When called about a rising creatinine in a cirrhotic patient with ascites:

1. Stop offenders — hold diuretics, NSAIDs, nephrotoxins, ACEi/ARBs; hold beta-blockers if MAP below 65 mmHg.
2. Urine studies — Na, Cr (calculate FeNa), sediment. If on diuretics → send FeUrea. Bland sediment with FeNa below 0.2% points toward HRS.
3. Rule out infection — blood cultures, diagnostic paracentesis if ascites present. SBP is the single most common precipitant.
4. Volume-assess clinically — if genuinely hypovolaemic, give albumin 1 g/kg (max 100 g). If euvolaemic or overloaded, do not give albumin to satisfy a protocol.
5. Reassess at 24 hours — if creatinine has not improved and no alternative cause is found, this is probable HRS-AKI → start vasoconstrictors (norepinephrine in Canada) and arrange monitored care.
6. Transplant — flag for urgent transplant evaluation if eligible.

The albumin challenge exists to rule out simple prerenal AKI before committing to an HRS diagnosis. Its duration (24 vs 48 hours) is genuinely contested — see the diagnosis section below.

---

Functional kidney failure driven by splanchnic vasodilation and systemic inflammation in a patient with portal hypertension. The useful mental shorthand is AKI on portal hypertension, not AKI on cirrhosis — a patient with compensated cirrhosis and no ascites cannot have HRS. It accounts for roughly a fifth of AKI presentations in cirrhosis, and untreated it kills within weeks.

Two things have shifted recently and are worth holding onto: the diagnosis is no longer conceived of as pure exclusion, and the old ritual of a mandatory 48-hour albumin challenge before treating is now contested rather than obligatory. Liver transplantation remains the only cure. In Canada, norepinephrine is the practical first-line agent because terlipressin is not marketed here.

---

pathophysiology

Portal hypertension drives nitric oxide release in the splanchnic bed, and blood pools there. Effective arterial blood volume falls even though total body water is high, so the kidney reads hypovolaemia and switches on RAAS, the sympathetic nervous system, and vasopressin. The result is intense renal vasoconstriction and a falling GFR in structurally normal kidneys.

Two modifiers make this worse. Systemic inflammation from bacterial translocation and circulating PAMPs causes microcirculatory dysfunction and some genuine tubular injury — which is why HRS and acute tubular injury are not mutually exclusive. And cirrhotic cardiomyopathy leaves the heart unable to raise cardiac output under stress, so the compensatory response that would normally protect renal perfusion never arrives.

Common precipitants are infection (SBP above all), GI bleeding, alcohol-associated hepatitis, and rapid volume depletion from over-diuresis or diarrhoea.

---

diagnosis

terminology

“Type 1” and “type 2” are obsolete. Current terms are HRS-AKI (acute), HRS-AKD (acute kidney disease, under 90 days), and HRS-CKD (90 days or more).

it is not purely a diagnosis of exclusion

This is the conceptual shift in the 2024 ADQI/ICA consensus and the reviews that followed it. HRS-AKI can coexist with acute tubular injury and can arise on a background of established CKD. Waiting for a pristine, single-cause picture before treating delays therapy in patients who would have responded. Treat the HRS physiology you can see while investigating the rest.

baseline creatinine

In hierarchy of preference: the lowest stable value in the previous 3 months; failing that, the most recent stable value within 12 months; failing that, the lower of the admission creatinine or a back-calculated value using an eGFR of 75 mL/min/1.73 m².

criteria

All of the following:

1. Cirrhosis with ascites.
2. AKI by KDIGO — creatinine rise of 26.5 µmol/L or more within 48 h, or a 50% or greater rise from baseline within 7 days, and/or urine output at or below 0.5 mL/kg/h for 6 h or more. Treat the urine output criterion cautiously; baseline output is often low in cirrhosis.
3. No response to volume correction. Withdraw diuretics and, in patients who are genuinely hypovolaemic, give albumin 1 g/kg/day (max 100 g). Reassess at 24 hours. A patient who is clearly euvolaemic or overloaded should not be given empiric albumin simply to satisfy a protocol.
4. No alternative cause — no shock, no nephrotoxins, no structural markers (proteinuria above 500 mg/day, more than 50 RBC/hpf).

distinguishing it from the alternatives

FeNa is unreliable in anyone on diuretics — use FeUrea (below 35% suggests a prerenal/HRS physiology) in that situation. Urinary NGAL is the biomarker with the best supporting data for separating HRS from acute tubular injury, with higher levels favouring tubular injury; availability is patchy and thresholds are not standardised, so it supports rather than replaces clinical judgement.

Two quick bedside clues. A normal or high-normal blood pressure argues against HRS, though bear in mind mean MAP in HRS cohorts sits around 76 mmHg and the comparison that matters is against the patient’s own baseline. And a normal or high sodium argues against it too — the ADH drive in HRS should be producing hyponatraemia.

---

prevention

Albumin has two well-supported prophylactic indications: SBP (1.5 g/kg on day 1, 1 g/kg on day 3) and large-volume paracentesis (6–8 g per litre removed when more than 5 L comes off).

It is not indicated for non-SBP infections, where it carries a risk of pulmonary oedema without a survival benefit, and it should not be given simply to push the serum albumin above 30 g/L.

---

management

Start treatment promptly and titrate to a MAP rise of 10–15 mmHg, or to a MAP above 65 mmHg. Roughly 40–50% of patients achieve HRS reversal with a vasoconstrictor plus albumin.

norepinephrine — first-line in Canada

Continuous infusion of 0.5–3 mg/hr, titrated to the MAP target. It has generally been given in an ICU or step-down setting with central access, though the practical bar is monitored care with reliable vascular access rather than a hard requirement for a central line before starting.

Meta-analyses (Olson, PLoS One. 2024; Pitre, Crit Care Med. 2022) support non-inferiority to terlipressin for HRS reversal, with a lower rate of respiratory failure. The mechanistic reason matters and explains most of the difference between the two drugs: norepinephrine’s β1 inotropy lets the heart compensate for the afterload it imposes, whereas terlipressin is a pure V1 agonist with no inotropic support at all. Terlipressin simultaneously raises afterload and shifts splanchnic blood centrally, raising preload, in a patient whose ventricle may already be unable to respond — and it may cause direct pulmonary venoconstriction on top of that.

terlipressin — the international standard

Gold standard where it is available, on the strength of CONFIRM (NEJM 2021). Not marketed in Canada; access requires a Special Access Programme application, so it is rarely the practical choice here. FDA-approved in the US since 2022. (Worth re-checking Canadian status periodically — this is the kind of thing that changes.)

Dosing runs two ways. European practice favours continuous infusion, starting at 2 mg/day and rising to a maximum of 12 mg/day, which appears to reduce ischaemic side effects. The North American label uses bolus dosing at 0.85 mg q6h, escalated to 1.7 mg q6h if creatinine has not fallen by at least 30% by day 4.

Contraindications: hypoxia with SpO₂ below 90%, ACLF grade 3 (three or more organ failures), creatinine above 440 µmol/L, and severe ischaemic vascular disease. Beyond the well-known ischaemic and respiratory risks, a 2025 pooled analysis of the three North American phase III trials identified a meaningful rate of bradycardia and other arrhythmias — worth continuous cardiac monitoring, and a newer signal than most references carry.

Albumin alongside a vasoconstrictor is usually 20–50 g/day (AASLD 2021 suggests 1 g/kg on day 1 then 40–50 g/day; the lower end of 20–40 g/day is often more appropriate depending on volume status).

midodrine plus octreotide

Midodrine 5–15 mg PO q8h with octreotide 100–200 µg SC q8h (or 50 µg/hr IV). This combination is substantially inferior to IV vasoconstrictors, and recent expert commentary argues it should largely be abandoned rather than treated as a legitimate alternative. Reasonable uses are narrow: bridging a patient awaiting ICU transfer, or treating someone for whom ICU admission has been judged inappropriate after a goals-of-care discussion. It is not a ward-based equivalent of norepinephrine.

when to stop

Stop for success (creatinine back within 26.5 µmol/L of baseline), for futility (no improvement after 4 days at maximum tolerated dose, per AASLD 2021 — assess response at day 4), for toxicity (ischaemia, pulmonary oedema, significant arrhythmia), or at 14 days maximum.

other interventions

TIPS is not recommended in HRS-AKI as a routine measure, given the risks of precipitating heart failure and hepatic ischaemia, though selected stable patients are occasionally considered at specialist centres. Beta-blockers should be held when MAP falls below 65 mmHg or AKI develops, and reassessed once the patient recovers.

---

transplant and prognosis

Liver transplantation is the definitive treatment, and every eligible patient with HRS-AKI should trigger urgent transplant evaluation. Whether a kidney is included — simultaneous liver-kidney, or SLK — turns on whether the renal failure is judged irreversible.

SLK criteria

These thresholds come from US OPTN policy. Canadian SLK allocation is set provincially and does not follow OPTN — check your own programme’s criteria rather than quoting these at a Canadian transplant meeting. They remain the most commonly examined framework and a reasonable conceptual anchor.

• HRS-CKD: eGFR at or below 60 mL/min for more than 90 days, plus either dialysis dependence or eGFR at or below 30 at listing.
• HRS-AKD: sustained AKI — dialysis or eGFR at or below 25 mL/min — for at least 6 consecutive weeks. The reasoning is that beyond about six weeks of profound renal failure, native recovery after liver transplant alone becomes unlikely.
• Safety net: a patient who misses the 6-week cutoff, receives a liver alone, and is still on dialysis or has an eGFR below 20 between 60 and 365 days post-transplant gets priority access to a kidney.

prognosis outside transplant

For a patient who is not a transplant candidate, the outlook is poor enough that palliative discussions belong early. Dialysis typically buys 3–6 months and is often a bridge to nowhere; without it, survival is measured in days to weeks. Even vasoconstrictor responders have high short-term mortality if transplant is off the table. In practice, severe HRS-AKI often comes with blood pressure too low to tolerate haemodialysis at all, which settles the question on technical grounds.

follow-up

The goal after discharge is catching the transition from reversible HRS-AKI to permanent HRS-CKD. Reassess kidney function within a month. Use cystatin C where you can — creatinine systematically overestimates GFR in sarcopenic cirrhotic patients.

---

related: norepinephrine · terlipressin · midodrine · spontaneous bacterial peritonitis · cirrhosis · acute kidney injury