Volume 16, Issue 9 pp. 2516-2531
Minireview
Free Access

Protecting the Kidney in Liver Transplant Candidates: Practice-Based Recommendations From the American Society of Transplantation Liver and Intestine Community of Practice

J. G. O'Leary

Corresponding Author

J. G. O'Leary

Division of Hepatology, Baylor University Medical Center, Dallas, TX

Co-first authors.Corresponding author: Jacqueline G. O'Leary, [email protected]Search for more papers by this author
J. Levitsky

J. Levitsky

Division of Gastroenterology and Hepatology, Department of Medicine, Northwestern University Feinberg School of Medicine, Chicago, IL

Co-first authors.Search for more papers by this author
F. Wong

F. Wong

Division of Gastroenterology, Department of Medicine, Toronto General Hospital, University Health Network, University of Toronto, Ontario, Canada

Search for more papers by this author
M. K. Nadim

M. K. Nadim

Division of Nephology and Hypertension, Department of Medicine, University of Southern California, Los Angeles, CA

Search for more papers by this author
M. Charlton

M. Charlton

Intermountain Transplant Center, Murray, UT

Search for more papers by this author
W. R. Kim

W. R. Kim

Division of Gastroenterology, Department of Medicine, Stanford University, Stanford, CA

Search for more papers by this author
First published: 17 March 2016
Citations: 52

See also: J. Levitsky et al.

Abstract

Acute kidney injury (AKI) and chronic kidney disease (CKD) are common in patients awaiting liver transplantation, and both have a marked impact on the perioperative and long-term morbidity and mortality of liver transplant recipients. Consequently, we reviewed the epidemiology of AKI and CKD in patients with end-stage liver disease, highlighted strategies to prevent and manage AKI, evaluated the changing liver transplant waiting list's impact on kidney function, delineated important considerations in simultaneous liver–kidney transplant selection, and projected possible future transplant policy changes and outcomes. This review was assembled by experts in the field and endorsed by the American Society of Transplantation Liver and Intestinal Community of Practice and Board of Directors and provides practice-based recommendations for preservation of kidney function in patients with end-stage liver disease.

Abbreviations

  • ACS
  • acute coronary syndrome
  • ADQI
  • Acute Dialysis Quality Initiative
  • AIN
  • acute interstitial nephritis
  • AKI
  • acute kidney injury
  • AKIN
  • Acute Kidney Injury Network
  • ATN
  • acute tubular necrosis
  • BPH
  • benign prostatic hyperplasia
  • CKD
  • chronic kidney disease
  • CNI
  • calcineurin inhibitor
  • DAA
  • direct-acting antiviral
  • ESRD
  • end-stage renal disease
  • GN
  • glomerulonephritis
  • HCV
  • hepatitis C virus
  • h
  • hour
  • HRS-1
  • type 1 hepatorenal syndrome
  • HRS-2
  • type 2 hepatorenal syndrome
  • HRS
  • hepatorenal syndrome
  • ICA
  • International Club of Ascites
  • ICU
  • intensive care unit
  • IV
  • intravenous
  • KDIGO
  • Kidney Disease Improving Global Outcomes
  • KIM-1
  • kidney injury molecule 1
  • LT
  • liver transplant
  • MAP
  • mean arterial pressure
  • MELD
  • Model for End-Stage Liver Disease
  • NASH
  • nonalcoholic steatohepatitis
  • NGAL
  • neutrophil gelatinase-associated lipocalin
  • NSAID
  • nonsteroidal anti-inflammatory drug
  • OPTN
  • Organ Procurement and Transplantation Network
  • RRT
  • renal replacement therapy
  • SBP
  • spontaneous bacterial peritonitis
  • SCr
  • serum creatinine
  • SLKT
  • simultaneous liver–kidney transplant
  • SQ
  • subcutaneous
  • TID
  • three times per day
  • TIMP-1
  • tissue inhibitor of matrix metalloproteinase 1
  • TIPS
  • transjugular intrahepatic portosystemic shunt
  • UNOS
  • United Network for Organ Sharing
  • UO
  • urine output
  • Introduction

    Kidney function is an important predictor of morbidity and mortality in patients with various chronic medical illnesses. In liver transplant (LT) candidates, kidney dysfunction is both common and associated with serious consequences, including increased mortality risk both before and after LT. Post-LT kidney function is determined by at least three factors: (i) pretransplant kidney injury, (ii) perioperative damage and recovery in kidney function, and (iii) posttransplant kidney injury. The first factor is addressed in this paper, and the remaining two are addressed in our companion paper 1. Preservation of kidney function in patients with end-stage liver disease (ESLD), through prevention of acute kidney injury (AKI), accurate diagnosis of AKI and chronic kidney disease (CKD), and timely institution of therapy, are critical determinants of post-LT outcomes. We reviewed the science and graded the available data on preserving pre-LT kidney function. All authors reviewed the data available, and recommendations were graded according to the Grades of Recommendation Assessment, Development & Evaluation (GRADE) system (Table S1) 2, 3.

    Epidemiology, Pathophysiology, and Diagnosis of Acute Kidney Injury and CKD in Patients With ESLD

    Kidney dysfunction is a common complication of liver cirrhosis, especially in patients with ascites. In patients with ESLD, this occurs in ≈20% of hospitalized patients 4 and in >50% of outpatients with decompensated cirrhosis 5. Most cases of kidney dysfunction in advanced cirrhosis are related to AKI; however, CKD is becoming more prevalent as increasing numbers of patients with chronic liver disease develop diabetes 6. Consequently, the spectrum of kidney dysfunction in cirrhosis is broad, including AKI occurring over days and CKD occurring over months to years. Most causes of kidney dysfunction in advanced cirrhosis are related to functional hemodynamic changes resulting from cirrhosis and may reverse with hemodynamic correction 4. Examples include volume depletion, resulting in prerenal azotemia that is responsive to diuretic withdrawal and volume expansion, and kidney dysfunction unresponsive to volume expansion, such as hepatorenal syndrome (HRS). Type 1 HRS (HRS-1) occurs over days, and type 2 HRS (HRS-2) is a less acute process, occurring over weeks to months. Kidney dysfunction related to structural disease (e.g. acute tubular necrosis (ATN), acute interstitial nephritis or glomerular diseases) is less reversible without disease-specific therapies.

    The major hemodynamic abnormality underlying functional kidney alterations in cirrhosis is splanchnic and systemic vasodilatation, which causes effective arterial underfilling. This leads to compensatory vasoconstrictor system activation (renin–angiotensin–aldosterone and sympathetic nervous systems) and results in kidney vasoconstriction that first conserves sodium and then water and finally reduces kidney blood flow to a level that impairs GFR 7. This decreased perfusion sets the stage for further kidney hemodynamic compromise from additional insults. The most common insult that precipitates kidney failure is bacterial infection 8. Bacterial products and the cytokines they induce have vasodilatory properties that promote further splanchnic and systemic vasodilatation, thereby worsening circulatory dysfunction. Bacterial products also can alter kidney peritubular microcirculation, inflict kidney damage directly and cause oxidative stress 9, 10, which in turn affects cellular metabolism and induces apoptosis.

    The definition of AKI in cirrhosis has recently undergone significant changes. Practically, patients with HRS-1 were previously not considered candidates for treatment unless their serum creatinine (SCr) reached ≥2.5 mg/dL over a period of ≤2 weeks; however, data have emerged suggesting that smaller acute changes in SCr also negatively affect patient outcomes 11. This led to the development of various diagnostic criteria for acute renal failure: RIFLE (risk, injury, failure, loss of function and ESRD) 12, the Acute Kidney Injury Network (AKIN) 13 and Kidney Disease Improving Global Outcomes 14. These various definitions of AKI have been proposed in the general population and are based on relative changes in SCr (rather than a threshold of SCr), urine output or initiation of renal replacement therapy (RRT) (Table 1) 13, 15. In 2010, the Acute Dialysis Quality Initiative, along with several members of the International Club of Ascites (ICA), recommended adaptation of the AKIN criteria to define AKI in patients with cirrhosis instead of the traditional definition using a fixed SCr cutoff value of >1.5 mg/dL (Table 1) 16, 17. These criteria were adapted regardless of the cause of AKI. As such, HRS-1 was categorized as a specific type of AKI. Since then, the use of the AKIN criteria in predicting mortality has been validated in several studies of hospitalized patients with cirrhosis including those in intensive care units (ICU) 5, 18-20. SCr was chosen as the measure of kidney function despite its shortcomings because it is familiar and readily available to all clinicians. Calculated GFR in a population with a high prevalence of sarcopenia, such as in cirrhosis, can lead to a falsely increased GFR, and urine output can be misleading because of avid sodium and water retention that both increase the risk of inaccuracies. The same small change in SCr (≥0.3 mg/dL) was adopted to define AKI in cirrhosis without staging (Table S2). CKD was also defined based on the National Kidney Foundation definitions. Functionally, this change allowed patients with acute deterioration of kidney function on a background of CKD to also be better defined.

    Table 1. Definition and staging of acute kidney injury
    AKI definition AKI stage AKI stage
    SCr criteria UO criteria
    1 2 3 1 2 3
    AKIN (2007) 13 Increase SCr ≥ 0.3 mg/dL (26.5 μmol/L) within 48 h; or increase SCr ≥1.5× baseline within 48 h; or UO <0.5 ml/kg/h × 6 h Increase ≥0.3 mg/dL (>26.5 μmol/l) within 48 h or ≥1.5–2× baseline Increase 2–3× baseline Increase 3× baseline or SCr > 4 mg/dL (>354 μmol/L) with an acute rise >0.5 mg/dL (44 μmol/L) or on RRT <0.5 mL/kg/h × 6–12 h <0.5 mL/kg/h × 12 h <0.3 mL/kg/h × 24 h or anuria × 12 h
    Baseline SCr is first SCr measured
    KDIGO (2012) 14 Increase SCr ≥0.3 mg/dL (26.5 μmol/L) within 48 h; or increase SCr ≥1.5× baseline, which is known or presumed to have occurred within the prior 7 days; or UO <0.5 mL/kg/h for 6 h Increase ≥0.3 mg/dL (>26.5 μmol/L) within 48 h or ≥1.5–2× baseline Increase 2–3× baseline Increase 3× baseline or SCr > 4 mg/dl (>354 μmol/L) with an acute rise >0.5 mg/dL (44 μmol/L) or on RRT <0.5 mL/kg/h × 6–12 h <0.5 mL/kg/h × 12 h <0.3 mL/kg/h × 24 h or anuria × 12 h
    Unknown baseline SCr estimation based on the MDRD formula, assuming a normal GFR ≈75–100 mL/min/1.73 m2
    ADQI (2010) 16, 17; AKI in cirrhosis Increase SCr ≥0.3 mg/dL (26.5 μmol/L) within 48 h or increase SCr ≥1.5× baseline Increase ≥0.3 mg/dL (>26.5 μmol/l) within 48 h or Increase 2–3× baseline Increase 3× baseline or
    HRS-1 is a specific form of AKI ≥1.5–2× baseline SCr >4 mg/dL (>354 μmol/L) with an acute rise >0.5 mg/dL (44 μmol/L) or on RRT
    ICA (2015) 21; AKI in cirrhosis Increase SCr ≥0.3 mg/dL (≥26.5 μmol/L) within 48 h or increase SCr ≥50% from baseline which is known or presumed to have occurred within the prior 7 days Increase ≥0.3 mg/dL (>26.5 μmol/L) within 48 h or ≥1.5–2× baseline Increase 2–3× baseline Increase 3× baseline or SCr >4 mg/dL (>354 μmol/L) with an acute rise >0.5 mg/dL (44 μmol/L) or on RRT
    SCr within 3 months can be used as baseline; in patients with more than one SCr value, value closest to hospital admission should be used; in patients without previous SCr, SCr on admission should be used
    • Adapted from Nadim et al 124. AKI, acute kidney injury; AKIN, Acute Kidney Injury Network; ADQI, Acute Dialysis Quality Initiative; h, hour; HRS-1, type 1 hepatorenal syndrome; ICA, International Club of Ascites; KDIGO, Kidney Disease Improving Global Outcomes; RRT, renal replacement therapy; SCr, serum creatinine; UO, urine output.

    More recently, the ICA 21 defined how baseline kidney function should be determined and outlined criteria for AKI progression, regression and treatment response (Table 1). HRS-1 was renamed HRS-AKI, with removal of the SCr cutoff value of ≥2.5 mg/dL, and was defined as development of stage ≥2 AKI (regardless of final SCr) provided all the other HRS-1 diagnostic criteria are fulfilled. Once validated, it is hoped that these revised criteria will facilitate research collaborations, standardize study protocols, enable earlier therapeutic initiation and reinvigorate pharmaceutical and academic investment in novel treatment strategies.

    Although oliguria is not included in the current definition of AKI in patients with cirrhosis, urine output has been found to be a sensitive and early marker for AKI in ICU patients and is associated with adverse outcomes 22-24.

    Key Points and Recommendations

    • AKI, defined as an increase in baseline SCr of ≥0.3 mg/dL within 48 h, is the preferred means of categorizing AKI in patients with cirrhosis (grade 1B).
    • HRS-AKI is a subset of AKI in patients with cirrhosis (grade 1B).
    • More than half of patients with decompensated cirrhosis have at least some degree of kidney dysfunction (grade 1C).

    Prevention of AKI

    Prevention of AKI is critical because it is associated with increased mortality in patients with cirrhosis 25 and is one of the most powerful predictors of post-LT survival. Several, often coexistent physiological and clinical insults can negatively affect kidney function (Table 2). More recently, the role of the gut microbiome translocation of bacteria or bacterial products and systemic proinflammatory responses have been increasingly recognized as important contributors in the pathogenesis of organ dysfunction, including AKI and CKD 26. As such, long-term spontaneous bacterial peritonitis (SBP) prophylaxis with daily antibiotics 27-29, intravenous (IV) albumin use in patients with SBP 30, 31, prompt fluid replacement of gastrointestinal blood loss together with antibiotic prophylaxis for gastrointestinal bleeding 32, 33, and simultaneous administration of IV albumin with large volume paracentesis (>5 L) 34 have all been shown to decrease the probability of HRS-AKI. Future interventional strategies may target altering the microbiome, preventing bacterial translocation or abrogating proinflammatory responses. Although rare, abdominal compartment syndrome (defined as increased intra-abdominal pressure to >20 mmHg) usually from tense ascites may lead to AKI by increasing venous pressure and causing arterial vasoconstriction 35, which may improve following paracentesis with IV albumin 36, 37. Drugs may exert a direct nephrotoxic effect by several mechanisms: (i) direct kidney tubule toxicity (e.g. radiocontrast dye, aminoglycosides, vancomycin, amphotericin B); (ii) allergic interstitial injury (e.g. nonsteroidal anti-inflammatory drugs [NSAIDs], β-lactam antibiotics, diuretics); and (iii) intrakidney blood flow impairment (e.g. radiocontrast dye, NSAIDs, renin–angiotensin–aldosterone system blockers) 38. In addition, changes in drug distribution due to volume overload and portal hypertension and altered pharmacokinetics due to changes in kidney and hepatic blood flow and function can clinically significantly modulate the concentration and half-life of medications and their metabolites.

    Table 2. Prevention of AKI in patients with cirrhosis
    Risk factors Preventive strategies
    Hepatorenal syndrome development • Antibiotic prophylaxis following gastrointestinal bleeding for 7 days (grade 1A)
    • Albumin infusion during large-volume paracentesis (>5 L, 6–8 g/L of ascitic fluid removed) (grade 1A)
    • Secondary and primary SBP prophylaxis with daily antibiotics, preferably norfloxacin (grade 1B)
    • Early recognition and treatment of SBP with antibiotics and IV albumin at the dose of 1.5 g/kg of body weight at the time of diagnosis of SBP and 1 g/kg of body weight on the third day of treatment (grade 1B)
    • Judicious use of diuretics
    • Avoid dehydration with lactulose use
    Exposure to nephrotoxic medications (e.g. NSAIDs, aminogyclosides, amphotericin, vancomycin) • Close monitoring of drug toxicity and early recognition of drug-induced AKI and discontinuation of offending agent if possible (grade 1A)
    • Use lipid formulations of amphotericin B rather than conventional formulations of amphotericin B (grade 2A)
    • Use azole antifungal agents and/or the echinocandins rather than conventional amphotericin B, if equal therapeutic efficacy can be assumed (grade 1A)
    • Avoid nephrotoxic medications whenever possible
    Radiocontrast exposure • Consider alternative imaging methods or avoidance of IV contrast if possible
    • Use low or iso-osmolar agents with lowest volume possible (grade 1B)
    • Optimize fluid status prior to administration of IV contrast with IV normal saline or IV bicarbonate (grade 1A)
    • Consider N-acetylcysteine use in combination with IV hydration (grade 2D)
    Hemodynamic instability • Increase mean arterial pressure in setting of shock to >65 mmHg (grade 1C)
    • Use of protocol-based management of hemodynamic and oxygenation parameters (grade 2C)
    • Optimal fluid resuscitation with crystalloids or colloids (grade 2B)
    • Vasopressors in patients with persistent hypotension (grade 1C), consider norepinephrine as first line (grade 2D)
    • AKI, acute kidney injury; IV, intravenous; NSAID, nonsteroidal anti-inflammatory drug; SBP, spontaneous bacterial peritonitis.

    Key Points and Recommendations:

    • AKI is best prevented in cirrhotic patients through utilization of SBP prophylaxis, IV albumin use in patients with SBP, antibiotic prophylaxis for gastrointestinal bleeding, and simultaneous administration of IV albumin with large volume paracentesis (grade 1A).

    Management of AKI

    Diagnosing the etiology of AKI is critical in determining therapy (Figure 1). Regardless of the AKI etiology, removing potential precipitating factors such as diuretics and optimizing volume status should be initiated even before a cause of AKI is established (Table 3). Intravascular volume expansion is an important part of treatment but also part of establishing the AKI etiology. Patients in whom other causes of AKI have been ruled out should receive treatment for HRS with vasoconstrictors in combination with IV albumin (Table 3) 39-57. Meta-analyses have shown a correlation between vasoconstrictor therapy increasing mean arterial pressure, improvement in kidney function and short-term survival benefit, regardless of the agent used 58, 59. Terlipressin is the most extensively studied and documented first-line therapy for HRS-AKI in countries with access; however, because of the unavailability of terlipressin in North America, midodrine in combination with octreotide 39, 43, 54 and, in particular, norepinephrine 49, 51, 54, 55 have also been shown to be beneficial in the treatment of HRS 40-42, 44-48, 50, 52, 53, 56 (Table 4). Unlike midodrine/octreotide 56, norepinephrine has been shown to be equivalent to terlipressin, although the expense of norepinephrine use, the need for ICU monitoring, and the inferior quality of data continue to support terlipressin as the agent of choice in areas with access 60. The optimal duration of medical treatment is not well established, although patients who failed to demonstrate improvement in SCr after day 4 are less likely to have HRS reversal 61, and an indication for treatment beyond 14 days has not been established. Increase in mean arterial pressure ≥5 mmHg during treatment, baseline SCr <3 mg/dL, Child–Pugh score <13, lower Model for End-Stage Liver Disease (MELD) score, younger age and bilirubin <10 mg/dL have all been shown to be independent predictors of therapeutic response. Of note, recent data have compared infusion and bolus therapy with terlipressin and shown similar responses to both regimens, with fewer side effects and a lower total daily dose needed in those receiving infusional therapy 62. Ideally, those patients with response to therapy would be given priority for transplant to avoid the need for simultaneous liver–kidney transplant (SLKT) 63.

    Details are in the caption following the image
    Algorithm for management of AKI. ACS, acute coronary syndrome; AIN, acute interstitial nephritis; AKI, acute kidney injury; ATN, acute tubular necrosis; BPH, benign prostatic hyperplasia; GN, glomerulonephritis; HRS, hepatorenal syndrome.
    Table 3. Management of AKI
    Therapies Comments Recommendations
    (Dosing and mechanisms of action)
    IV fluids Crystalloid or colloids
    • Diarrhea or overdiuresis requires crystalloids (grade 1A)
    • Gastrointestinal bleeding needs packed red blood cells if hemoglobin is <7 g/dL (grade 1A)
    • IV fluids should be administered carefully due to risk of volume overload (grade 1C)
    • IV medication concentration or conversion to oral medication if possible
    RRT Hemodialysis or continuous RRT
    • RRT should be initiated in patients with worsening AKI, fluid overload despite diuretic therapy or problematic acid-base status (grade 1D)
    Treatment of HRS-1
    Terlipressin Vasopressin analog
    • HRS-1 patients need resuscitation with IV albumin (initially 1 g/kg for 2 days, up to a maximum of 100 g/day, followed by 20–40 g/day) in combination with a vasoconstrictor (grade 1A), preferably terlipressin (grade 1A)
    0.5–1.0 mg every 4–6 h, increased on day 4 to 2 mg IV every 4–6 h if SCr has not decreased by >30% from baseline following fluid resuscitation; IV infusion of terlipressin is an alternative to bolus dosing with similar efficacy and fewer side effects
    • In countries in which terlipressin is not available, norepinephrine can be used as an initial therapy in ICU patients or as an alternative in patients in whom midodrine and octreotide has failed (grade 1C)
    Midodrine plus octreotide α-Adrenergic agonist (midodrine)
    • We recommend TIPS in patients with HRS-1 with lower MELD scores (grade 1C)
    Somatostatin analog (octreotide)
    • Artificial liver support therapies for HRS should be limited to research only (grade 2D)
    7.5 mg orally TID with increase to 12.5–15 mg TID to increase MAP by 15 mmHg plus octreotide SQ 100 μg TID, titrated to 200 μg TID on day 2, if renal function has not improved
    Norepinephrine α-Adrenergic agonist
    0.5–3.0 mg/h continuous IV infusion to increase MAP by 10 mmHg
    TIPS Decreases portal hypertension with subsequent decrease in vasoconstriction mediators (vasopressin, norepinephrine, endothelin and angiotensin II) leading to improved cardiac output and GFR
    Albumin dialysis Dialyzing blood against albumin containing solution across a highly permeable high-flux membrane; blood-bound toxins are cleared by diffusion and taken up by binding sites of the albumin dialysate
    • AKI, acute kidney injury; HRS, hepatorenal syndrome; HRS-1, type 1 hepatorenal syndrome; ICU, intensive care unit; IV, intravenous; MAP, mean arterial pressure; RRT, renal replacement therapy; SQ, subcutaneous; TID, three times per day; TIPS, transjugular intrahepatic portosystemic shunt.
    • a Given in combination with intravenous albumin (initial dose 1 g/kg followed by 20–40 g/day).
    • b In Europe, continuous infusion is used rather than intermittent doses of terlipressin because lower doses can be used with similar results and fewer side effects.
    • c The doses quoted for midodrine were initially reported in a study of five patients. There is no absolute rule to raise the MAP by 15 mmHg, rather the MAP has to be raised to an adequate level to provide renal perfusion. The use of SQ octreotide is an alternative to continuous infusion.
    Table 4. Results of studies using vasoconstrictor therapy in patients with HRS-1
    Author Year Study design Patients Patients with HRS-1, % Treatment Albumin HRS reversal, %
    Angeli 39 1999 Retrospective 5 100 Midodrine plus octreotide No 100
    8 Dopamine 0
    Colle 40 2002 Retrospective 11 100 Terlipressin Yes/no 60
    Halimi 41 2002 Retrospective, multicenter 18 89 Terlipressin No 72
    Moreau 42 2002 Retrospective, multicenter 99 100 Terlipressin Yes 58
    Esrailian 43 2007 Retrospective 60 100 Midodrine plus octreotide No 40
    21 No treatment 10
    Uriz 44 2000 Prospective 9 67 Terlipressin Yes 77
    Ortega 46 2002 Prospective 13 76 Terlipressin Yes 77
    8 Terlipressin No 25
    Duvoux 47 2002 Prospective 12 100 Norepinephrine Yes 83
    Solanki 48 2003 Randomized 12 100 Terlipressin Yes 42
    12 Placebo 0
    Alessandria 49 2007 Randomized 12 41 Terlipressin Yes 83
    10 Norepinephrine 70
    Neri 50 2008 Randomized, multicenter 26 100 Terlipressin Yes 81
    26 Placebo 19
    Sharma 51 2008 Randomized 20 100 Terlipressin Yes 50
    20 Norepinephrine 50
    Sanyal 52 2008 Randomized, multicenter 56 100 Terlipressin Yes 34
    56 Placebo 13
    Martin-Llahi 53 2008 Randomized, multicenter 23 56 Terlipressin Yes 44
    23 Placebo 9
    Singh 55 2012 Randomized 23 100 Terlipressin Yes 39
    23 Norepinephrine 43
    Cavallin 56 2015 Randomized 27 92 Terlipressin Yes 70
    22 Midodrine plus octreotide 29
    Cavallin 62 2016 Randomized 39 100 Terlipressin infusion Yes 56
    39 Terlipressin bolus 46
    Boyer 57 2016 Randomized 97 100 Terlipressin Yes 24
    99 Placebo 15
    • HRS, hepatorenal syndrome; HRS-1, type 1 hepatorenal syndrome.

    Transjugular intrahepatic portosystemic shunt (TIPS) has been shown to improve kidney function in small studies of patients with HRS 64-66. TIPS, however, is contraindicated in patients with severe hepatic dysfunction, defined as serum bilirubin >5 mg/dL, a high MELD score, significant kidney dysfunction, cardiac failure or clinically significant hepatic encephalopathy.

    The initiation of RRT should be made on clinical grounds, including hyperkalemia, oliguria with volume overload, metabolic acidosis, refractory hyponatremia not responding to medical management, and diuretic resistance or intolerance. Optimal timing for RRT indication has not been studied in patients with cirrhosis; however, data from AKI studies in critically ill patients without liver disease suggest that early RRT initiation and maintenance of negative fluid balance in those with volume overload may improve survival 67-71. In patients receiving RRT, continuous RRT allows for the slower correction of serum sodium and provides greater cardiovascular stability compared with standard hemodialysis 72, 73.

    Although initial small studies of albumin dialysis using the Molecular Adsorbent Recirculating System (MARS) documented a survival benefit compared with standard medical therapy 74, a more recent multicenter study (RELIEF trial) failed to reproduce this benefit at 28 days, despite biochemical improvement 75.

    Key Points and Recommendations

    • Diuretic withdrawal and volume replacement are essential to determining etiology of AKI (grade 1A).
    • Terlipressin is the first-line therapy for HRS-AKI in countries with access to this medication (grade 1A).

    Impact of AKI Treatment on Patient Survival and Kidney Function After LT

    The cumulative rates of stage 4 or 5 CKD and ESRD at 10 years after LT have been shown to be 18% and 25%, respectively 76. Although these high rates may be partially attributable to long-term calcineurin inhibitor (CNI) use, diabetes and hypertension, the most important predictor of CKD after LT is pretransplant kidney function. The development and duration of AKI before transplant not only reduces post-LT kidney function long term but also increases mortality 77-86. Reversal of HRS-1 with vasoconstrictors before LT has been associated with improved post-LT outcomes 8, 87, 88. At the present time, specific pre-LT predictors of CKD development after LT are lacking.

    Key Points and Recommendations

    Key points and recommendations are outlined in Table 3.

    Evolution in the composition of the LT waitlist

    The composition of the LT waitlist in the United States and Europe is being transformed by three events that are increasing the rate of pre- and posttransplant AKI and CKD. First is the decades-long obesity epidemic, with a resulting parallel increase in the prevalence of nonalcoholic steatohepatitis (NASH). Second is the development of highly effective and well-tolerated direct-acting antiviral (DAA) agents for hepatitis C virus (HCV) infection. Third is adoption of the MELD score, which includes SCr as the most heavily weighted value in MELD-based organ allocation.

    In addition to the contribution of ESLD to kidney dysfunction, some causes of cirrhosis have disease-specific kidney injury mechanisms. HCV, for example, is associated with membranoproliferative glomerulonephritis and type 2 cryoglobulinemia. Reversal of kidney dysfunction is not an inevitable consequence of HCV eradication, even for cryoglobulinemia 89. Similar to HCV infection, NASH is independently associated with CKD 90, 91 and alcoholic liver disease, which is associated with IgA nephropathy.

    Liver disease secondary to HCV infection continues to be the most common indication for LT, affecting approximately one-third of LT recipients 92. NASH is the third most common indication for LT in the United States and the second most common reason for being waitlisted 93. This may still be an underestimate of NASH frequency because it does not account for the likely high prevalence of NASH among the 10% of recipients with “cryptogenic” cirrhosis 93.

    Although the number of patients with HCV infection in need of LT is projected to rise in the coming decade 92, the rate of increase is a small fraction of that occurring for NASH 93. The frequency of HCV patients needing LT will also be affected by the advent of well-tolerated DAAs; recent clinical trials of DAAs have demonstrated safety and efficacy, even in patients with decompensated cirrhosis 94. In addition, HCV viremia has been associated with an increased risk of CKD and ESRD 95; although data are lacking, one would expect improved renal outcomes after virologic cure.

    Based on these well-established trends in the U.S. LT waitlist composition, it is highly probable that NASH and HCV will compose more than half of transplant activity for at least the next decade. Although the association of NASH with pre- and post-LT CKD is likely related to the relatively high prevalence of diabetes and hypertension, NASH is predictive of CKD independent of other factors 90, 96. Because both of these diseases are independently predictive of pre- and post-LT CKD, the rise in NASH frequency and the continued high prevalence of HCV as top indications for LT will almost certainly result in increased rates of CKD before and after LT.

    Implications of AKI Etiology on Mortality Risk of MELD Score

    The near ubiquitous presence of kidney hypoperfusion/ischemia in cirrhosis, with subsequent low fractional excretion of sodium, can make the diagnosis of ATN difficult. Robust data regarding the reversibility of AKI in patients with cirrhosis were generated in a prospective study of 562 patients with cirrhosis and AKI 97. The authors observed that HRS was less common than prerenal or infection-associated kidney injury. Of the nearly 500 patients in whom a diagnosis could be made, the most common precipitating causes of AKI were infection (46%), prerenal injury (32%), HRS (13%), and parenchymal kidney disease (9%). Approximately 80% of AKI in patients with cirrhosis is secondary to a treatable precipitating cause, and it has been estimated that 10–20% of patients with cirrhosis who develop AKI have an element of ATN.

    The curvilinear association with increasing MELD score and waitlist mortality, coupled with the direct relationship of MELD score with likelihood of undergoing LT, raises the possibility that treating AKI in patients on the waitlist may improve short-term mortality but may decrease the short-term likelihood of LT. The net effect of longer waitlist time, with associated increased waitlist mortality, together with longer duration of kidney dysfunction is difficult to calculate and has not been modeled. Perhaps partly for this reason, survival following HRS is primarily dependent on reversal of hepatic failure with less significant survival improvement following HRS reversal 52, 98. Instead, treating AKI (other than HRS-1) in waitlisted patients presumes reversibility of AKI and a net benefit of successful treatment.

    When weighing the relative merits of a specific organ offer, it is important to consider that if AKI contributed to a MELD score, 90-day mortality is differentially affected depending on the AKI etiology 97. Differential weighting of SCr according to AKI etiology could, in theory, be used to reduce waitlist mortality by utilizing ΔSCr instead of absolute SCr values 99.

    Key Points and Recommendations

    • The increased prevalence of NASH, the advent of DAA therapy for HCV infection and MELD-based organ allocation have changed the LT waiting list (grade 1A) and likely increased the prevalence of CKD before and after LT (grade 1C).
    • The etiology of kidney dysfunction in patients with ESLD alters its impact on prognosis (grade 1B).

    Considerations for SLKT

    Since the introduction of the MELD score to assign priority for donor allocation, the proportion of LT recipients that undergo SLKT has nearly tripled 100 The increase in SLKT frequency is a reflection of the increasing frequency of ESRD among LT recipients and the high priority for LT assigned to patients with kidney dysfunction. In addition to MELD-based allocation increasing the risk of AKI and CKD before LT, the MELD score at transplant is increasing nationwide 93. Consequently, kidney dysfunction has almost become a requirement for patients to advance to the top of the waitlist 101. At the time of LT, a high percentage of ESLD patients have significant AKI with or without CKD, with variable improvement after LT 77, 102-106. It is well known that recipients with AKI that results in prolonged dialysis after LT have reduced survival 107; therefore, it is no surprise that SLKT utilization has increased 108. This would be acceptable if kidney allografts were readily available, but there is already a severe kidney allograft shortage for patients with isolated ESRD who also have high waitlist mortality. Furthermore, the addition of a kidney transplant to the LT procedure adds significant operative time, potential morbidity and marked cost.

    Ideally, patients with HRS-AKI with a favorable response to vasoconstrictor therapy and intravenous albumin would receive prioritization for transplant with a liver only to facilitate a single-organ transplant before irreversible kidney dysfunction occurred 63. Alternatively, MELD coefficients could be changed to decrease the weight of kidney dysfunction in the MELD score; however, because changes to MELD have proven difficult, the creation of stringent criteria to access SLKT has been proposed. In 2008, a United Network for Organ Sharing (UNOS) consensus conference evaluated allocation of kidneys to LT candidates with kidney dysfunction 109. Several agreed-upon parameters predicting nonrecovery of native kidney function after LT included (i) CKD with estimated (MDRD equation) GFR ≤30 mL/min, (ii) CKD on kidney biopsy (defined as >30% glomerulosclerosis and/or 30% fibrosis), (iii) AKI with SCr ≥2.0 mg/dL and dialysis ≥8 weeks, (iv) special consideration of patients with comorbidities (e.g. hypertension and diabetes) and patients aged >65 years, and (v) metabolic kidney disease (e.g. hyperoxaluria, atypical hemolytic uremic syndrome, methylmalonic aciduria) 109. Conversely, not meeting these UNOS criteria implies an expectation of native kidney function recovery to acceptable levels after LT alone. The UNOS criteria, however, are not official Organ Procurement and Transplantation Network (OPTN) policy; therefore, the current allocation system allows listing for SLKT based on subjective clinical judgment, with the UNOS criteria serving as guidelines. A recent survey of U.S. transplant centers showed that AKI leading to dialysis for a minimum of 4 weeks and to CKD (defined mainly by GFR <30 mL/min) were minimal criteria used by most centers to recommend SLKT 110; however, because adherence to this guidance is not mandatory, SLKT selection criteria continue to vary dramatically across the United States 110. Unfortunately, neither center nor national guidelines predict kidney recovery with a high level of certainty 77, 109, 111, 112. Other predictors, such as kidney ultrasound, measured GFR (iothalamate, iohexol) and kidney biopsy, may be too subjective, inaccurate, invasive or costly for universal clinical implementation for SLKT selection.

    Ultimately, dissemination of appropriate, mandatory selection strategies is needed. Recently, the OPTN/UNOS Kidney Transplantation Committee in collaboration with the Liver Intestine Transplant Committee proposed more specific SLKT medical eligibility criteria with a safety net for LT-only allograft recipients with a continued need for kidney transplantation. Specifically, pre-LT patients with CKD would need documentation of dialysis or GFR ≤35 mL/min, and patients with AKI would need GFR ≤25 mL/min for ≥6 consecutive weeks prior to SLKT listing 113. For patients who do not meet these criteria and who do not recover kidney function after LT alone, kidney transplant waitlist prioritization has been proposed. Prioritization would occur after OPTN review verified GFR ≤20 mL/min or continued dialysis 60–365 days after LT. If accepted, these criteria would provide greater stringency for SLKT allocation and would facilitate kidney transplant in LT recipients with persistent kidney dysfunction after liver-only transplantation. Fortunately, a similar policy supporting stringent criteria for SLKT allocation and the potential for a safety net for kidney allograft allocation in LT recipients without renal recovery has been proposed in Canada 114.

    Even with specific clinical criteria for SLKT allocation, preoperative biomarker assessment would, ideally, allow more accurate kidney allograft allocation (Table 5) 111. Recent data have demonstrated that biomarkers can more appropriately classify, quantitate and prognosticate kidney dysfunction 18, 115, 116. The majority of these markers, such as kidney injury molecule 1 and neutrophil gelatinase-associated lipocalin, are elevated in acute tubular injury, contrast nephrotoxicity or perioperative kidney injury but may also be markers of kidney disease progression 117. Other plasma protein profiles (osteopontin, tissue inhibitor of metalloproteinase 1) before LT in conjunction with clinical variables (age, diabetes) might be able to better predict post-LT kidney function recovery but need clinical decision algorithm testing for SLKT versus LT alone 118. Furthermore, pre-LT biomarker assessment may facilitate early post-LT institution of kidney-sparing immunosuppression regimens. Although these data require further validation, future studies should test the utility of serum and urine biomarkers as serial measures to guide management, including the decision to perform SLKT or to implement nephroprotective strategies. This might have a greater potential to assist individualized therapeutic and organ-utilization strategies if combined with immunological risk-assessment biomarkers.

    Table 5. Biomarkers of acute and chronic kidney injury in patients with cirrhosis and after LT
    Biomarker Site of origin Source Renal injury in cirrhosis? Renal injury in LT recipients?
    NGAL (or lipocalin-2) 116, 125-130 Activated neutrophils, renal tubule Serum, urine AKI AKI
    Cystatin-C 115, 131-136 All nucleated cells Serum, plasma AKI/CKD AKI/CKD
    KIM-1 127, 137 Renal tubule Serum, urine, tissue AKI
    IL-8, IL-18 126, 127, 129, 137 Macrophages, epithelium Urine AKI AKI
    Apolipoproteins AI, H, and CIII 115, 138 Hepatocyte, enterocyte Serum CKD
    α1, β2 microglobulins 115, 139 α1 (Liver), β2 (all nucleated cells) Serum CKD CKD
    TIMP-1 118 Renal tubule, many tissues Plasma AKI
    Toll-like receptor 4 140 Macrophages, dendritic cells, fibroblasts Urine AKI
    Trefoil factor 3 115, 141 Enterocyte, epithelium AKI CKD
    TGF-β1 142, 143 Lymphocytes, myeloid cells Urine CKD CKD
    l-Fatty acid binding protein 1 115, 118, 127, 130 Hepatocyte, enterocyte Urine AKI AKI/CKD
    Prostaglandin D2 synthase 144, 145 Mast cells Serum, plasma, urine CKD CKD
    Osteopontin 118, 141 Fibroblasts, osteoblasts, most immune cells, muscle, endothelium, kidney Serum, plasma, urine AKI
    • AKI, acute kidney injury; CKD, chronic kidney disease; KIM-1, kidney injury molecule 1; LT, liver transplant; NGAL, neutrophil gelatinase-associated lipocalin; TIMP-1, tissue inhibitor of matrix metalloproteinase 1.

    Key Points and Recommendations

    • Using current criteria to predict which patients have irreversible kidney failure requiring SLKT versus reversible kidney failure requiring LT alone is suboptimal (grade 1C).
    • National criteria for SLKT allocation combined with a safety net of kidney transplant prioritization for LT-only recipients with postoperative kidney failure could improve kidney organ allocation in the ESLD population (grade 1D).
    • Preoperative biomarkers of kidney recovery that would allow more specificity in decision making for SLKT versus LT only have the potential to improve kidney organ allocation in the ESLD population (grade 1C).

    Looking to the Future

    An essential step in optimizing post-LT GFR is better understanding and management of pre-LT kidney dysfunction. When attempting to determine a target minimum GFR after LT, recent data indicate that the relation between post-LT kidney function and subsequent mortality may be accurately quantified. In a large single-center study, a >30% reduction in estimated GFR between 3 and 12 mo after LT was associated with increased mortality (odds ratio 2.6; p < 0.001) 119. In a separate study, a GFR 15–29 mL/min had a hazard ratio of 2.7 (p < 0.01), and a GFR <15 mL/min had a hazard ratio of 5.5 (p < 0.01) for mortality 120.

    CNI nephrotoxicity remains a major contributor to kidney injury following LT. Future CNI-mitigation strategies may include individualized management of immunosuppression. Genetic polymorphisms such as cytochrome P450 34A, ATP-binding cassette subfamily B-1 and nitric oxide synthase 3, for example, have been linked to susceptibility to CNI toxicity 121-123 and could be tested before transplant to plan for posttransplant immunosuppression. Further validation of these and other associations accompanied by functional confirmation may inform personalized application of immunosuppression. For more information on optimizing post-LT renal function, please see our companion paper 1.

    At the policy level, the best information must shape future allocation decisions. SLKT utilization and equitable/appropriate access to kidney transplantation via a safety net is likely to remain contentious, but a consensus is urgently needed. To maximize post-LT survival outcome, an optimal policy regarding SLKT would aim for a post-LT GFR target of >60 mL/min in the greatest number of organ recipients. In addition, introduction of novel therapeutic agents would require adaptation of the organ allocation policy. Although future therapy for HRS, for example, may restore SCr, whether the corresponding reduction in the MELD score is commensurate with mortality benefits based on kidney functional recovery must be studied. Finally, promoting timely organ allocation and preventing pre- and postoperative morbidity and mortality would require prospective multicenter data collaborations and accurate risk assessment tools to inform data-driven interventions and rational policy development addressing all stakeholders in LT.

    Key Points and Recommendations

    • The relationship between functional impairment of the kidney and subsequent mortality may be assessed quantitatively, with GFR <60 mL/min portending higher mortality risk (grade 2C). Future research on how to mitigate kidney injury in the pre- and post-LT periods is needed to achieve optimal post-LT kidney function.
    • Future organ allocation policy must incorporate best available scientific data and adopt rational approaches to optimize pre- and posttransplant outcomes for all organ transplant patients (i.e. liver alone, kidney alone and SLKT) (grade 2C).

    Acknowledgments

    We acknowledge the support of the American Society for Transplantation Board of Directors and the Liver and Intestine Community of Practice.

      Disclosure

      The authors of this manuscript have conflicts of interest to disclose as described by the American Journal of Transplantation. J.G.O. has received honoraria from Gilead, Abbvie, Astellas, Novartis, grant funding from Grifols and Fisher Scientific and is a consultant for Gilead, Abbvie, Grifols, and Intercept. J.L. has received honoraria from Novartis, Gilead, Salix, grant funding from Novartis, and is a consultant for Transplant Genomics Inc. M.N. is a consultant for Baxter and Mallinckrodt. M.C. has received research support from Gilead, Merck, Janssen, Abbvie, Novartis, Galectin, and Intercept and is a consultant for Gilead, Bristol Myers, Merck, Janssen, Abbvie, Novartis, Galectin, and Intercept. F.W. is a consultant for Mallinckrodt. W.R.K. has no conflicts of interest to disclose.

        The full text of this article hosted at iucr.org is unavailable due to technical difficulties.