Introduction

Hepatorenal syndrome (HRS) is a multisystem disorder occurring in those with advanced liver disease. HRS presents in decompensated liver cirrhosis with an incidence of 4%. The chance of HRS increases from 18% in the first year to nearly 40% in the fifth year. ¹ Pathophysiology of HRS presents in the setting of cirrhosis inciting a neurohormonal changes that result in splanchnic and systemic vasodilation ultimately leading to renal hypoperfusion resulting in reflex renal vasoconstriction culminating in renal failure. ² HRS can be characterized by acute kidney injury (AKI) with a high mortality (HRS-AKI), or by progressive renal function decline and refractory ascites with a better prognosis (HRS-non-AKI).³

HRS-AKI has a worse prognosis compared to HRS-non-AKI with only 10% of patients surviving longer than 90 days.⁴ Diagnostic criteria for HRS-AKI include presence of (1) decompensated liver cirrhosis and ascites (2) AKI (3) no response of AKI after 2 days of volume expansion and holding diuretics (4) absence of shock and nephrotoxic agents (5) no evidence of structural kidney injury.⁵ HRS-AKI (formerly known as Type 1 HRS) can be triggered by spontaneous bacterial peritonitis, gastrointestinal hemorrhage, major surgery or large volume paracentesis. Early diagnosis of HRS can allow for a better prognosis.⁶ 

Treatment of HRS-AKI includes prompt discontinuation of nephrotoxic medications, antihypertensives, diuretics and nonsteroidal anti-inflammatory drugs followed by initiation of vasoconstrictors and albumin. Vasoconstrictor options include terlipressin, norepinephrine or midodrine.⁷ Reversal of HRS-AKI is defined by improved renal function achieving a complete response.⁸Review of the literature identifies factors that predict the chance of response of HRS-AKI to the above treatment modalities; however, there are only a few published studies primarily focusing on terlipressin, a therapy that is not universally available in the United States and was not approved by the FDA until 2022. These studies have shown that a higher change in mean arterial pressures, lower baseline bilirubin and creatinine levels, presence of systemic inflammatory response syndrome and lower urine neutrophil gelatinase-associated lipocalin (NGAL) were associated with response to terlipressin and albumin.³ Studies evaluating predictive factors of HRS-AKI response to albumin plus norepinephrine or midodrine and octreotide are lacking.

This is a retrospective study performed prior to the FDA approval of terlipressin. Our aim was to identify factors predictive of response of HRS-AKI to albumin plus midodrine and octreotide. Identifying key lab values or other patient characteristics at the time of admission or during hospitalization can help to tailor treatment and establish an individualized management plan. This data may help select patients who would benefit from standard treatment or require more aggressive care. Furthermore, data from this study may identify patients who will not benefit from therapy allowing for more effective allocation of resources influencing hospital expenditure on treatment of HRS-AKI reducing treatment cost.    

Methods and Data Analysis

A retrospective chart review was performed on 371 patients who were admitted with a primary diagnosis of decompensated cirrhosis and acute kidney injury per ICD-9/10 codes. Per International Club of Ascites (ICA) guidelines patients were diagnosed with AKI or non-AKI HRS⁵.

Patients meeting ICA criteria for a diagnosis for HRS-AKI were included. Thereafter, the following inclusion criteria was established for our study: cirrhosis with ascites, an acute increase in Cr to greater than 2.5 mg/dL within a two-week span, absence of shock defined as systolic blood pressure (SBP) < 90 mmHg, no recent exposure to nephrotoxic agents, the absence of proteinuria (>500 g/dl) or pre-existing parenchymal disease, and no improvement in creatinine after discontinuation of diuretics and starting volume expansion. Exclusion criteria include cardiac, respiratory failure, cancer, or prior beta-blocker use for variceal bleeding. (Figure 1).

Figure 1: Flow diagram of patient enrollment

All patients were treated with albumin, midodrine, and octreotide. Following 3 days of therapy, patients were classified as demonstrating a complete response – decrease in Cr to < 1.5, partial response – decrease in Cr by > 50% from baseline but not to < 1.5, no response – no decrease or increase in Cr by day 3 of treatment. Treatment was continued beyond 3 days up to 14 days if the patient’s Cr improved with therapy. Total bilirubin, albumin, and INR at admission and peak creatinine were compared among the different responder groups. Cutoff values for HRS-AKI reversal were expressed as mean values for complete response. 

Statistics: 

Categorical outcomes were evaluated using Fisher’s exact or Chi-squared test, as appropriate. Categorical variables were expressed as percentages. Continuous parametric variables were evaluated with the two-tailed Student’s t-test assuming unequal variance and nonparametric variables with Mann–Whitney U test. Normally distributed continuous variables were described as means and SD while skewed data were described as medians and interquartile ranges. Data analysis was conducted using SAS JMP Pro 13 software (SAS Institute Inc., Cary, North Carolina, USA).

Results

78 patients met inclusion criteria of which 23 (29.49%) achieved a complete response, 8 (10.26%) had partial response, 47 (60.26%) were non responders. In complete responders 65.2% were male with an average age 59.0 ± 8.6, MELD-Na 28.5 ± 5.2, predominant cause of cirrhosis being alcohol 52.2%, and 52.2% of patients demonstrated >180-day survival. In partial responders 62.5% were male with an average age 63.8 ± 6.4, MELD-Na 28.1 ± 9.9, predominant causes of cirrhosis being alcohol 50.0% and HCV 50.0%, and 62.5% of patients demonstrated < 30-day survival. In non-responders 63.8% were male with an average age 56.6 ± 10.3, MELD-Na 28.7 ± 8.7, predominant causes of cirrhosis being alcohol 38.3% and HCV 23.4%, and 46.8% of patients demonstrated < 30-day survival. (Table 1).

Table 1: Demographics 
For complete, partial, and non-responders, the mean total bilirubin (mg/dL) in the respective groups were 6.27±4.72, 9.76±8.47 and 10.94 ±11.78. The mean albumin (g/dL) in each respective group were 3.04±0.60, 2.74±0.66 and 2.71±0.50. Mean INR were 1.90±0.91, 1.81±0.40, and 1.91±0.88. Mean Peak creatinine were 3.17±1.72, 3.24±0.84, and 4.05±1.61.