Hepatorenal syndrome is the kidney paying for the liver's failures. The parenchyma is innocent — transplant these kidneys into a healthy recipient and they work — but the circulation that cirrhosis creates strangles their perfusion. That single fact, that HRS is functional, drives everything: the diagnosis is one of exclusion, the treatment aims at the circulation rather than the kidney, and the definitive cure is a new liver. Get the label wrong, in either direction, and you either withhold a treatment that works or pour vasoconstrictors into an ATN that needs supportive care.
The wrecked circulation: why the kidney vasoconstricts
Portal hypertension drives the release of nitric oxide and other vasodilators in the splanchnic bed, so blood pools in a dilated splanchnic circulation. The effective arterial blood volume — what the baroreceptors sense — falls, even as total body water rises into ascites and oedema. The body responds as it would to any perceived underfilling: the renin-angiotensin system, the sympathetic nervous system, and vasopressin all fire, clamping down on systemic and especially renal vessels. The kidney, caught in this vasoconstriction, sees its perfusion collapse and its GFR fall. It is the same functional logic as pre-renal AKI, with one cruel difference: the underfilling cannot be corrected with fluid, because the problem is a leaking, dilated splanchnic reservoir, not an empty tank.
Two further insults compound it. Cirrhotic cardiomyopathy blunts the heart's ability to raise its output to compensate, so the circulation cannot rescue itself. And systemic inflammation — driven by bacterial translocation from the gut and the molecular debris of a failing liver — adds a non-haemodynamic injury that helps explain why infection so reliably tips a patient into HRS. The modern picture is haemodynamic and inflammatory, not haemodynamic alone.
A name that changed to start treatment sooner
The old split into type 1 (rapid, with creatinine doubling past 2.5 mg/dL within two weeks) and type 2 (slower, with refractory ascites) has been retired. The problem was the threshold: waiting for creatinine to reach 2.5 meant waiting while a salvageable kidney deteriorated. The current scheme diagnoses HRS-AKI on the same AKI criteria used everywhere else — a rise of 0.3 mg/dL in 48 hours or 1.5 times baseline in 7 days — so therapy can begin at a lower creatinine. The slower, non-AKI forms are now HRS-NAKI, subdivided into HRS-AKD and HRS-CKD. The renaming is not pedantry; it is a deliberate lowering of the treatment threshold.
Diagnosis by exclusion: the criteria and the challenge
HRS has no confirmatory test; it is what remains after the alternatives are excluded. The ICA criteria require cirrhosis with ascites, AKI by the standard criteria, and no improvement after two days of withdrawing diuretics and expanding volume with albumin at 1 g/kg/day. They also require the absence of shock, no recent nephrotoxins or contrast, and no sign of structural injury — minimal proteinuria, no significant microhaematuria, and a normal renal ultrasound. The albumin challenge sits at the centre of this: a hypovolaemic pre-renal kidney, given enough albumin, recovers; the HRS kidney does not, because its underfilling is not a volume deficit you can fill. The challenge is therefore diagnostic and therapeutic in the same gesture.
The differential in a cirrhotic with AKI
AKI is common in cirrhosis and HRS is only one cause, so the differential must stay open. Hypovolaemic pre-renal AKI — from over-diuresis, lactulose-driven diarrhoea, bleeding, or large-volume paracentesis without albumin — is the great mimic, and it is the one the albumin challenge is designed to catch. ATN from sepsis, shock, or nephrotoxins is the other major alternative, and it must not be dressed up as HRS, because it needs supportive care rather than vasoconstrictors. Acute interstitial nephritis and the glomerular diseases that accompany liver disease — IgA nephropathy, or cryoglobulinaemic disease in hepatitis C — round out the list, flagged by an active sediment. The classic indices disappoint here: FENa is often below 0.1% in HRS, but it can also be low in ATN complicating cirrhosis, so it cannot reliably separate the two. Urinary damage biomarkers such as NGAL run higher in ATN than HRS and may help, but they are not yet decisive.
Treating the circulation: vasoconstrictor plus albumin
Because the lesion is a dilated splanchnic circulation with renal vasoconstriction, treatment reverses both: a vasoconstrictor to tighten the splanchnic bed and raise effective volume, and albumin to expand the circulation and bind inflammatory mediators. Terlipressin, a vasopressin analogue, is first-line where it is available and reverses HRS-AKI in roughly four to five in ten treated; a continuous infusion appears to cause fewer adverse events than boluses. Its harms are real — ischaemic events and, importantly, respiratory failure — so it is avoided in advanced acute-on-chronic liver failure, hypoxaemia, and significant coronary disease. Noradrenaline plus albumin is a comparably effective alternative in the ICU, cheaper and familiar, needing a central line and monitoring. The midodrine-octreotide-albumin combination is weaker and reserved for settings where neither terlipressin nor an ICU is available. Albumin runs throughout: around 1 g/kg on day one, then roughly 20 to 40 g daily, watching for the pulmonary oedema that over-expansion can cause.
None of this works in isolation from the precipitant. Spontaneous bacterial peritonitis must be treated promptly with antibiotics, and albumin given alongside it — 1.5 g/kg on day one and 1 g/kg on day three — prevents HRS in its own right. Bleeding must be controlled, nephrotoxins stopped, and over-diuresis reversed.
The definitive answer, and the bridges to it
Liver transplantation is the only treatment that addresses the cause, and because the kidneys are functionally rather than structurally injured, they usually recover once a working liver restores the circulation — so most patients do not need a combined liver-kidney transplant unless the renal injury has been prolonged or is structural. Everything else is a bridge or an adjunct. A transjugular intrahepatic portosystemic shunt lowers portal pressure and helps selected patients, limited by encephalopathy and cardiac reserve. Renal replacement therapy bridges transplant candidates and those with a reversible precipitant; offered as a destination to a patient who is neither a transplant candidate nor recovering, it raises a goals-of-care question this volume answers in its decision chapters rather than here.
Where the evidence is firm, and where it argues
The mechanism and the nomenclature are settled, and the principle of vasoconstrictor-plus-albumin is supported by randomised trials. The arguments are at the edges. Terlipressin's trials show it reverses HRS-AKI more often than placebo or midodrine-octreotide, but the same data carry a real signal of respiratory failure that has shaped how and to whom it is given. The indices that textbooks once trusted — FENa above all — perform poorly in cirrhosis, and the biomarkers meant to replace them are promising but not yet definitive. And the hardest questions, around dialysis and transplant candidacy, are as much about values and prognosis as about physiology. The chapter grades these honestly rather than pretending the uncertainty away.
