Abstract
Patients with liver disease often have concomitant kidney disease. It is often difficult to recognize or is underestimated, as use of serum creatinine or creatinine-based equations to estimate glomerular filtration rate is inaccurate in patients with cirrhosis. This chapter discusses the impact of kidney dysfunction on the prognosis of patients with liver disease. It discusses the etiology, treatment, and prognosis of the various types of acute kidney injury (AKI) that develop in patients with cirrhosis, the most serious of which is hepatorenal syndrome (HRS). In addition, the diagnosis, etiology, and treatment of intrinsic kidney disease that occurs in conjunction with or as a result of liver disease are discussed. Finally, recovery of kidney function following successful liver transplantation and the indications for simultaneous liver-kidney transplantation are discussed.
Keywords
acute kidney injury, hepatic cirrhosis, creatinine-based equations, prerenal azotemia, abdominal compartment syndrome, terlipressin, posttransplant renal recovery, simultaneous liver-kidney transplantation
Assessing Kidney Function in Cirrhotic Patients
There is no reliable method to assess kidney function in patients with cirrhosis. Serum creatinine is unreliable (especially in white women) because of their low muscle mass and because of the reduced creatine (the precursor of creatinine) generation seen in cirrhosis. Bilirubin levels also interfere with the serum creatinine measurement assessed by the Jaffé method. Because of these factors, for any given rise in serum creatinine, the glomerular filtration rate (GFR) drop is more pronounced in cirrhosis than the general population with normal muscle mass. Studies have demonstrated that almost a third of cirrhotic patients with a serum creatinine that falls in the normal range have a GFR less than 50 mL/min per 1.73 m 2 . Creatinine-based equations used to estimate GFR, such as the Modification of Diet in Renal Disease (MDRD) and the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI), also overestimate GFR in the majority of patients because of the limitations of serum creatinine measurement, especially at lower levels of true GFR. Twenty-four-hour creatinine clearance is cumbersome and overestimates GFR because of increased tubular secretion of creatinine in cirrhotic patients. Cystatin C is a low-molecular-weight protein that is freely filtered by the glomerulus and subsequently metabolized in the proximal tubules. Cystatin C generation is independent of age, sex, muscle mass, or bilirubin level, which makes it an attractive biomarker to assess kidney function in cirrhosis. However, concerns regarding assay variability, interaction of cystatin C with multiple drugs, and lack of familiarity with the test have limited its use. Although the accuracy of iohexol or iothalamate clearance is lower in patients with cirrhosis than in the general population (because ascites interferes with assessment of the volume of distribution [Vd] of the radio-labeled marker), these methods are considered to be the most accurate for determining kidney function in cirrhosis.
Acute Kidney Injury in Cirrhosis: Prevalence and Causes
Kidney dysfunction is common in patients with acute liver failure and those with cirrhosis. Studies have demonstrated that up to 50% of patients will develop one or more episodes of acute kidney injury (AKI) during the course of their illness. The risk of AKI also progressively increases with the severity of liver disease. Almost 20% of hospitalized patients with cirrhosis develop AKI during their incident hospitalization. The most common cause of AKI is volume-responsive prerenal kidney failure from diarrhea or excessive diuretic use, which accounts for almost 45% of cases. Acute tubular necrosis (ATN), from prolonged hypotension (e.g., following gastrointestinal bleeding) or exposure to nephrotoxic medications, and hepatorenal syndrome (HRS) is responsible for 32% and 23% of AKI episodes, respectively. Abdominal compartment syndrome from tense ascites ( Fig. 30.1 ) is an increasingly recognized cause of AKI in cirrhosis. Diagnosis is based on detecting an intravesical pressure of greater than 20 mm Hg, associated with evidence of AKI. The mechanism of kidney injury is poorly understood but seems to be related to renal vein compression with subsequent decline in renal venous return, renal blood flow (RBF), and GFR. Reduction of intraabdominal pressure with paracentesis results in a brisk diuresis, increase in urine flow, and improvement in serum creatinine. Other causes of AKI, including acute interstitial nephritis and obstructive uropathy, are rare in patients with cirrhosis; obstructive uropathy accounts for less than 1% of AKI in cirrhosis.
Definition of Acute Kidney Injury in Cirrhosis
AKI in cirrhosis has been traditionally defined as a rise in serum creatinine of 50% to a level ≥1.5 mg/dL (133 µmol/L). The 1.5 mg/dL cutoff is totally arbitrary and has not been validated in any prospective study. Acute Kidney Injury Network (AKIN) criteria, which depend on percent or absolute rise in serum creatinine as well as amount of urine output, correlate with hospital and out-of-hospital mortality in patients who develop AKI in the general population. Recent studies demonstrated that AKIN criteria also predicted in-hospital as well as 6-month out-of-hospital mortality in critically ill cirrhotic patients admitted to the intensive care unit. The AKIN criteria are also better correlated with 90-day mortality than a rise of serum creatinine ≥1.5 mg/dL in patients with advanced cirrhosis. Even patients with AKIN stage 1 and a peak serum creatinine less than 1.5 mg/dL had higher mortality than those who never developed AKI. Patients presenting with higher AKIN stage (stage 2 vs. stage 1 and stage 3 vs. stage 2) and those who progressed from one AKIN stage to the next demonstrated progressively worse survival than those who presented with AKIN stage 1 and whose kidney function recovered or stabilized without progression to a higher stage. In 2015, the International Club of Ascites (ICA) published a new definition for AKI in cirrhosis that aligns with the AKIN criteria for the general population. However, the urine output criterion was omitted from the ICA criteria because cirrhotic patients either are taking large doses of diuretics that affect the urine volume or are oliguric ( Table 30.1 ).
| Subject | Definition | |
|---|---|---|
| Baseline sCr |
| |
| Definition of AKI |
| |
| Staging of AKI |
| |
| Progression of AKI | Progression | Regression |
| Progression of AKI to higher stage and/or need for KRT | Regression of AKI to a lower stage | |
| Response to treatment | No Response | Partial Response | Full Response |
| No regression of AKI | Regression of AKI stage with a reduction of sCr to ≥0.3 mg/dL (26.5 µmol/L) above the baseline value | Return of sCr to a value within 0.3 mg/dL (26.5 µmol/L) of the baseline value |
HRS is a functional form of AKI that develops in cirrhotic patients with ascites and does not respond to volume expansion. For many years, type-1 HRS has been defined as rapid deterioration of kidney function over less than 2 weeks duration to a creatinine greater than 2.5 mg/dL, while type-2 HRS was defined as a gradual increase in serum creatinine to ≥1.5 mg/dL over weeks or months. The ICA has recently modified the definition of HRS to omit these serum creatinine cutoffs ( Table 30.2 ). Pathophysiology and treatment of HRS are discussed later in this chapter.
|
Identifying the Cause of Acute Kidney Injury in Cirrhosis
Identifying the cause of AKI in cirrhotic patients can sometimes be a difficult task. Improvement of kidney function with intravascular volume expansion with albumin (1 g/kg per day) will identify patients with prerenal kidney failure from volume depletion. An intravesical pressure greater than 20 mm Hg in the setting of AKI is diagnostic of abdominal compartment syndrome and should promote measures to reduce intraabdominal pressure. Kidney ultrasound is not essential in all cases but can be used if obstructive kidney failure is suspected. The presence of granular urinary casts, proteinuria greater than 500 mg/day, or microscopic hematuria exclude HRS. Classic teaching has advocated using fractional excretion of sodium (FENa) to differentiate between HRS and ATN, but recent studies have demonstrated that although urinary sodium excretion and FENa were lowest in patients with prerenal kidney failure, these urinary indices were also low in patients with ATN and did not differentiate between prerenal and intrinsic causes in cirrhotic patients with AKI. Because of the high risk of bleeding, kidney biopsy is rarely pursued to determine the etiology of AKI. However, there is no absolute contraindication to perform a kidney biopsy in these patients if coagulation parameters are normal, and findings from biopsy may be useful because previous studies demonstrate lack of correlation between histologic and clinical findings in cirrhosis patients. Levels of novel urinary biomarkers, including urinary neutrophil gelatinase-associated lipocalin (NGAL), are higher in patients with ATN compared with those without AKI, those with AKI from HRS, or those with prerenal kidney failure from volume depletion. However, the urinary NGAL values still show significant overlap between different causes of AKI, which limits the diagnostic accuracy of these assays. Other novel urinary biomarkers such as interleukin-18 (IL-18), kidney injury molecule-1 (KIM-1), and liver-type fatty acid-binding protein (L-FABP) are also higher in patients with ATN, but levels still overlap with those from patients with HRS. Urinary albumin excretion is also higher in patients with AKI from ATN, compared with patients with HRS, but urinary albumin excretion again shows significant overlap between these two diseases. Fig. 30.2 represents a general schema for differentiating and managing cirrhotic patients with AKI according to the AKIN stage at presentation.
Effect of Acute Kidney Injury on Survival
AKI has a strong impact on mortality in cirrhotic patients. Even after resolution of the AKI episode, patients who experienced AKI had worse survival compared with those who never developed AKI. The worst survival has been reported in patients with type-1 HRS, with some studies showing 2-week mortality as high as 80% in untreated patients. The introduction of vasoconstrictor therapy for HRS treatment and better patient care have improved survival, with one multicenter study reporting 90-day survival of 20% and 40% for patients with type-1 and type-2 HRS, respectively. Patients with volume responsive AKI and those with ATN still demonstrate lower survival compared with those without AKI episode. As alluded to earlier, progression to a higher AKI stage is associated with a progressive increase in mortality compared with those who recovered kidney function and those who did not progress to higher AKI stage.
Pathophysiology of Hepatorenal Syndrome
Patients with cirrhosis progress into different clinical stages. With early cirrhosis, there is increase in renal salt and water retention without any clinical evidence of ascites or peripheral edema (preascitic stage). With cirrhosis progression and further increase in renal sodium absorption, ascites starts to develop. Ascites is initially responsive to diuretics (diuretic-dependent ascites stage), but with subsequent progression of the liver disease, ascites becomes resistant to diuretics (diuretic-resistant or paracentesis-dependent ascites stage). This is the stage that precedes type-2 HRS development.
HRS is the result of a complex interplay between different pathways, including arterial vasodilatation, imbalance between vasodilator and vasoconstrictor substances, failure of renal autoregulation, systemic inflammatory response (SIRS), cardiac dysfunction, and adrenal insufficiency.
Arterial vasodilatation is the main pathophysiologic derangement that leads to HRS. Hepatic cirrhosis is associated with increased resistance of blood flow through the liver, which leads to portal hypertension, opening of portosystemic shunts, and preferential pooling of the blood volume in the splanchnic circulation. Increased production of various vasodilator mediators (e.g., nitrous oxide [NO]) that escape hepatic deactivation due to poor hepatic function and through portosystemic shunting reaches the systemic circulation, causing arterial vasodilatation, decreased systemic vascular resistance, decreased effective arterial blood volume, and hypotension. This results in off-loading of the baroreceptors in the carotid body and aortic arch and subsequent increased sympathetic nervous system (SNS) activity, increased circulating level of norepinephrine, tachycardia, and hyperdynamic circulation, with increases in cardiac output (CO). Despite the overall drop in systemic vascular resistance, vasoconstriction develops in localized vascular beds, including the brain and the kidney. In the kidney, relative arterial hypovolemia and renal vasoconstriction increase proximal sodium reabsorption and activate the renin-angiotensin-aldosterone system (RAAS), with subsequent increased aldosterone secretion, renal vasoconstriction, and salt and water retention. Nonosmotic release of vasopressin from the posterior pituitary gland helps maintain vascular tone and leads to increased free water absorption from the distal nephron and hyponatremia. In early cirrhosis (patients in the preascitic and diuretic-responsive ascites), the increase in the previously mentioned neurohumoral factors, along with the increase in CO, help maintain the effective circulating volume with no apparent change in kidney function. However, in late cirrhosis (diuretic-resistant ascites and beyond), these compensatory mechanisms fail to maintain an effective circulating volume, and renal vasoconstriction worsens, leading to HRS development. The hemodynamic and neurohumoral changes manifesting in cirrhotic patients in different stages of the liver disease are presented in Fig. 30.3 .
