Hepatorenal syndrome in the era of acute kidney injury

1 INTRODUCTION

Acute kidney injury (AKI) is a common complication occurring in patients with cirrhosis, with reported frequencies of up to 20%-50% in patients with cirrhosis admitted to hospital for complications of the disease.1, 2 Hepatorenal syndrome (HRS) is a unique form of AKI that develops in patients with advanced cirrhosis and is mainly related to a marked renal vasoconstriction secondary to the systemic circulatory impairment characteristic of patients with advanced cirrhosis.3 Although HRS is a unique form of AKI typical of patients with cirrhosis, these patients may develop AKI due to other different aetiologies.3, 4 The development of AKI is associated with impaired prognosis; however, prognosis markedly differs according to the cause of AKI.4 Therefore, early diagnosis and identification of the cause of AKI followed by the early administration of correct treatment is of utmost importance to improve outcomes.

Recently, the definition of AKI in cirrhosis has been modified according to the Acute Kidney Injury Network (AKIN) criteria that were already widely used for the definition of AKI in the general population.5 According to the new definition, the diagnosis of AKI is based on small changes in serum creatinine (SCr) with respect to previous values, rather than relying on a fixed cut-off value of SCr. The use of the new definition may lead to an earlier identification of episodes of renal impairment, which may have a beneficial impact on prognosis. This review will give an update on the diagnosis and management of HRS in cirrhosis in 2018, in the setting of the changes in AKI definition in cirrhosis and the emergent data on the potential usefulness of urine kidney biomarkers in this scenario.

2 NEW DEFINITION OF ACUTE KIDNEY INJURY IN CIRRHOSIS

Traditionally, kidney failure in cirrhosis has been defined as SCr > 1.5 mg/dL.6 Despite the definition being widely used for many years, there was evidence that it had limitations. Particularly, that it led to a delayed diagnosis of kidney failure, as the cut-off level identified patients who already had a marked decrease in glomerular filtration rate (GFR) (SCr > 1.5 mg/dL corresponds to approximately GFR < 30 mL/min)7; and that the use of a fixed cut-off level did not consider changes in SCr with respect to previous values.

In the last decade, the Acute Kidney Injury Network (AKIN) proposed new diagnostic criteria for kidney failure, currently named AKI, based on small changes in SCr.5 These criteria were validated in the general population and are now widely used in this setting. According to the AKIN criteria, AKI is defined as an increase in serum SCr ≥ 0.3 mg/dL occurring within a short period of time (48 hours).5 Moreover, AKI is classified into three different stages (AKI stage 1, 2 and 3) depending on the intensity of changes in SCr.

In recent years, several studies have investigated the usefulness of the AKIN criteria for the diagnosis of AKI in patients with cirrhosis.8-12 Results from these studies showed that the diagnosis of AKI, according to the new criteria and its stages, was good at prognostic stratification, as AKI was associated with increased in-hospital stay and mortality. On this background, in 2012 the International Club of Ascites (ICA) organized a consensus meeting to discuss a new definition of AKI in cirrhosis based on a modified version of the AKIN criteria. According to the new ICA-AKI criteria, AKI in cirrhosis is defined as an increase in SCr ≥ 0.3 mg/dL (≥26.5 μmol/L) within 48 hours; or a percentage increase in sCr ≥ 50% from baseline which is known, or presumed, to have occurred within 7 days.13 These criteria also stratify AKI into 3 different stages with respect to the intensity of the changes in SCr and provide definitions for the concepts of progression and regression of AKI, and response to treatment (Table 1).

An important point derived from the new definition is that it is essential to have a baseline value of SCr to apply the new diagnostic criteria. For patients developing AKI during hospitalization, this situation is easy as the SCr value at admission may be used as a baseline SCr.13 However, a high percentage of patients with cirrhosis develop AKI before hospitalization and, therefore, present with an already high SCr value at admission.2 In this population, the diagnosis of AKI should be made using pre-admission values of SCr as baseline. Patients with cirrhosis may develop impairment of kidney function during the progression of the disease; thus, the closest the baseline SCr value from admission, the more accurate the diagnosis. The ICA-AKI criteria arbitrarily defined baseline SCr as a value of SCr within the previous 3 months; if more than 1 value is available, the value closest to hospital admission should be used. In those patients without a previous SCr value, the SCr on admission should be used as baseline. It is important to highlight that for this latter group of patients who already have a high SCr at hospital admission but no baseline SCr is available, the diagnosis of AKI could be missed. Therefore, special attention should be made to this group of patients, especially if SCr ≥ 1.5 mg/dL. If a baseline SCr is not available, a formal diagnosis of AKI according to the ICA-AKI criteria cannot be made; however, clinical experience would suggests that patients with such an impairment of kidney function, especially if there is an identifiable precipitating event, probably have AKI and should be treated accordingly.

To overcome the problem of the absence of a baseline SCr in some patients, the proposal was made to use an estimated SCr calculated by the reverse application of the MDRD formula using a predefined value of GFR. However, MDRD underestimates GFR in patients with cirrhosis and, therefore, this method would be inaccurate and is not recommended in cirrhosis.13, 14

3 DIFFERENTIAL DIAGNOSIS OF AKI: ROLE OF URINE BIOMARKERS IN 2018

As mentioned before, HRS-AKI is a characteristic form of AKI that develops in patients with advanced cirrhosis.3 However, as a result of the impairment of circulatory function, patients with cirrhosis have labile kidney function and are prone to develop AKI due to aetiologies other than HRS, such as hypovolemia, ATN or nephrotoxicity. Differential diagnosis between different causes of AKI is essential as treatment completely differs between HRS-AKI and other causes of AKI.15 Moreover, prognosis differs according to AKI aetiology, with patients with hypovolemic AKI showing the best prognosis, compared to patients with HRS-AKI or ATN who have the worst prognosis.4

Currently, the differential diagnosis of the cause of AKI is based on clinical criteria, that include some degree of subjectivity, and standard analytical data. Classical urinary parameters such as proteinuria, fractional excretion of sodium (FeNa) and urine osmolality, which are used to help in the differential diagnosis, have several limitations, as they may be influenced by the use of diuretic treatment and have not been validated in large studies. In some cases, the differential diagnosis of the cause of AKI may easily be made only on clinical data with no need for further investigation. Nevertheless, some other clinical scenarios may be challenging, particularly the differential diagnosis between HRS-AKI and ATN, as these are both entities that occur in critically ill cirrhotic patients in whom several other complications may be present making the diagnosis more intricate.

Urinary biomarkers of tubular damage previously described in the nephrology field have shown to be potentially useful in the differential diagnosis of AKI in cirrhosis. Particularly, they have been shown to be accurate in distinguishing HRS-AKI from ATN. Among all these biomarkers, the most investigated and promising one appears to be neutrophil gelatinase-associated lipocalin (NGAL), followed by interleukin-18 (IL-18) and albumin.

NGAL is a glycoprotein that is overexpressed in the kidney by injured kidney tubular epithelia. Urinary NGAL (uNGAL) levels rise exponentially during AKI, prior to serum creatinine elevation.19 Several studies have shown that uNGAL levels are useful in the differential diagnosis of AKI in cirrhosis.20-23 AKI studies consistently show that patients with ATN have significantly higher uNGAL levels compared to those in patients with HRS-AKI or hypovolemia-induced AKI. Patients with hypovolemia show the lowest uNGAL levels, similar to those of patients without AKI. Interestingly, patients with HRS show intermediate uNGAL levels, which are significantly lower than those of patients with ATN and slightly higher than those of patients with hypovolemia-induced AKI. Interestingly, results from these studies show that uNGAL has good predictive accuracy for the differential diagnosis of ATN vs other causes of AKI with areas under the receiver-operating characteristic curves (AUROC) ≥ 0.8.20-23 It should be noted that uNGAL may increase in the context of urinary tract infections, as it is also produced by leucocytes. Therefore, uNGAL values should be interpreted with caution in patients with urinary tract infections.

Besides NGAL, other urine biomarkers, including IL-18, albumin and liver fatty acid binding protein (L-FABP), have also been investigated for the differential diagnosis between ATN and HRS.22, 23 Interestingly, the performance of all 3 biomarkers is similar, showing the highest values in patients with ATN, compared to patients with HRS-AKI or hypovolemia-induced AKI. Patients with HRS-AKI consistently show intermediate values between the other 2 groups. Among all these biomarkers, IL-18 and albuminuria have been the ones showing the best diagnostic accuracy in 2 studies.22, 23

Recently a meta-analysis investigated the role of uNGAL and urine IL-18 for the differential diagnosis and prognosis of AKI in cirrhosis.24 Results from this meta-analysis, including 8 prospective studies, showed that urinary levels of NGAL and IL-18 were able to discriminate between ATN and other types of AKI with high accuracy. The AUROC for the diagnosis of ATN was 0.86 (95%CI 0.68-0.94) for NGAL and 0.88 (95%CI 0.82-0.93) for IL18.24 Results from this meta-analysis and also from other individual studies show that biomarkers are not only useful for the differential diagnosis of AKI but can also be useful to predict short-term mortality. There are data showing that urinary levels of NGAL and IL-18 are significantly higher in patients who eventually died compared to those who were alive at the end of follow up.24

Overall, currently there is accumulating evidence showing that urine biomarkers and, particularly NGAL and IL-18, are useful for the differential diagnosis and to predict prognosis in patients with cirrhosis and AKI. In fact, the new EASL clinical practice guidelines on the management of decompensated cirrhosis recommend to implement the use of urine biomarkers, particularly NGAL, in the differential diagnosis between ATN and HRS-AKI.15 Nonetheless, the performance of these biomarkers in the setting of the new algorithm for the diagnosis of HRS-AKI in cirrhosis will require investigation in future studies.

4 PATHOPHYSIOLOGY OF HEPATORENAL SYNDROME-AKI

Great amount of evidence indicates that circulatory dysfunction plays a key role in the pathophysiology of HRS-AKI.3, 25 On this context, HRS-AKI has been traditionally defined as a characteristic cause of AKI of functional origin that develops in patients with cirrhosis, in the absence of histological kidney abnormalities. In recent years, there are data suggesting that the marked systemic inflammatory response present in patients with advanced cirrhosis may also be involved at some degree in the development of HRS-AKI.26

4.1 Systemic circulatory dysfunction

It is accepted that HRS-AKI occurs as a consequence of a marked reduction in renal blood flow and GFR secondary to impairment of effective arterial blood volume caused by splanchnic arterial vasodilation.3, 6, 25 The vasodilation of the splanchnic circulation is likely caused by the release of vasodilator factors, such as nitric oxide, carbon monoxide or endocannabinoids, as a consequence of portal hypertension.27-29

In early stages of cirrhosis, the increase in portal pressure is moderate and, therefore, there is slight reduction in systemic vascular resistance due to moderate splanchnic arterial vasodilation. At this stage of the disease, the reduction in systemic vascular resistance is compensated by an increase in cardiac output, which helps maintaining arterial pressure within normal values.3, 6, 25 In advanced stages of the disease, when patients have already developed decompensations, there is a marked reduction in systemic vascular resistance, as a result of a marked splanchnic vasodilation, that cannot be compensated by the increase in cardiac output.3 Moreover, there are data showing that at advanced disease stages, there is also a decrease in cardiac output that contributes to arterial underfilling.30 Overall, at this stage, patients develop a marked impairment of effective arterial blood volume. In this context and in order to maintain arterial pressure within normal limits, there is activation of endogenous vasoconstrictor systems, including renin-angiotensin system, sympathetic nervous system and, at later stages, a non-osmotic hypersecretion of vasopressin. These vasoconstrictor systems help maintain effective arterial blood volume and arterial pressure within normal limits, but have also important detrimental effects on kidney function. The effect of vasoconstrictor systems is associated with sodium and solute-free water retention, with the accumulation of ascites and oedema. If the activation of these systems is extreme, they lead to a marked renal vasoconstriction with the consequent reduction in GFR and development of HRS-AKI3, 6, 25 (Figure 1).

5 MANAGEMENT OF HEPATORENAL SYNDROME-AKI

The goal of the management of HRS-AKI, particularly in those patients awaiting liver transplantation (LT), is the normalization of kidney function in order to provide successful bridge to LT.39-42 Once diagnosed, HRS-AKI should be treated as soon as possible. Ideally, patients should be managed in an intensive or intermediate care unit, especially those awaiting LT. Patients should be closely monitored to assess kidney function evolution and for early detection of possible associated complications, particularly bacterial infections, and start treatment as soon as possible. Intravenous (IV) fluids should be administered with caution to prevent pulmonary oedema and the development/or further impairment of hypervolemic hyponatremia. The use of a central venous catheter is recommended to monitor central venous pressure in patients who are going to receive pharmacological treatment because such treatment involves the administration of albumin. The use of a bladder catheter is not necessary in all cases because it is associated with high risk of urinary tract infections. Therefore, its use is recommended only in patients with marked oliguria.41, 43

5.2 New algorithm for the management of AKI-HRS and implications for therapy

It is important to remark that according to the new diagnostic criteria of HRS-AKI a new algorithm for the management of AKI in patients with cirrhosis has been proposed (Figure 2, Table 2B). When AKI is diagnosed, the cause of AKI should be investigated as soon as possible, to prevent AKI progression. However, even in the absence of a definitive recognized cause, management of AKI should be immediately started according to the initial stage. Irrespective of the stage, diuretics and beta-blockers should be discontinued or tapered. Other precipitating factors of AKI should be identified and treated, including screening and treatment of infections, volume expansion when appropriate and discontinuation of all nephrotoxic drugs, such as vasodilators, beta-blockers or NSAIDs. Volume replacement should be used in accordance with the cause and the severity of fluid loss. Patients with diarrhoea, excessive diuresis or acute gastrointestinal bleeding should be treated with crystalloids or packed red blood cells respectively. In contrast, in all patients with AKI stage ≥ 1B, 20% albumin solution at a dose of 1 g of albumin/kg of body weight (with a maximum of 100 g of albumin) for 2 consecutive days should be administered. If patients do not respond to volume expansion with albumin and meet all HRS-AKI criteria, at this point treatment with vasoconstrictors and albumin should be started, if there are no contraindications for therapy.13, 15

It should be highlighted that using this algorithm the differential diagnosis and management of HRS will be made earlier as compared to when using classical type-1 HRS diagnostic criteria because the need to reach a cut-off level of SCr of 2.5 mg/dL has been removed.13 In this context, treatment with vasoconstrictors and albumin will be initiated with lower SCr values. Considering that the baseline SCr at the initiation of pharmacological therapy is a predictive factor of treatment response, this new diagnostic and treatment strategy should likely lead to a higher treatment response rate. However, to date there are no studies investigating the efficacy and the safety profile of the use of vasoconstrictors and albumin in patients with HRS using the new ICA-AKI criteria. Therefore, treatment should be ideally performed in the setting of prospective studies to assess efficacy and safety of such approach.

CONFLICT OF INTEREST

CS, EP and ES declare that they have no conflicts of interests. PG declares consultancy advisory boards for: Ferring Pharmaceuticals, Grifols, Novartis and Promethera. He has received research funding from Grifols S.A.