Burden and impact of acute kidney injury in acute liver failure: A systematic review

Abstract

BACKGROUND

Acute kidney injury (AKI) affects 40%–80% of patients with acute liver failure (ALF) and arises from complex mechanisms, including hemodynamic instability, drug toxicity, and direct renal insults. Unlike chronic liver disease, AKI in ALF is poorly studied.

AIM

To consolidate current evidence and guide future research and interventions to improve outcomes in ALF patients with AKI.

METHODS

This systematic review followed PRISMA guidelines, searching databases from 2000 to 2024 for studies on AKI in ALF. We included studies focused on adult, English-language, full-length epidemiological studies. Data were extracted on study design, patient demographics, risk factors, and outcomes. Methodological quality was assessed using a tailored 5-point tool. Data were summarized using descriptive statistics and analyzed with GraphPad Prism and MedCalc software.

RESULTS

This systematic review analyzed nine studies published between 2009 and 2024, encompassing 6031 patients with ALF, of whom 62.9% developed AKI. Most studies were retrospective and varied in methodology, setting, and population. Only three studies reported independent risk factors for the development of AKI among ALF patients, with significant variation in the risk factors studied. AKI in ALF patients significantly increased mortality, with hazard ratios ranging from 2.77 to 11.61. The review highlights the need for standardized diagnostic criteria and risk stratification to guide prevention and improve clinical outcomes in ALF.

CONCLUSION

AKI is common in ALF, increasing morbidity and mortality; standardised diagnosis and targeted management strategies are urgently needed.

Key Words: Acute liver failure; Acute kidney injury; Systematic review; Paracetamol related acute liver failure; Transient kidney injury; Persistent kidney injury

Core Tip: This systematic review of nine studies (6031 patients) found that acute kidney injury (AKI) occurs in 62.9% of acute liver failure (ALF) cases, with incidence varying from 37.9% to 79.1%. Drug-induced ALF, especially paracetamol-related, was the predominant etiology in most cohorts, while viral hepatitis was more common in Asia. Few studies identified independent AKI risk factors, with systemic inflammation, shock, and tumor necrosis factor-alpha elevation among them. AKI was consistently associated with increased mortality and lower transplant-free survival. Significant gaps include a lack of etiological data, pathophysiological insights, and representation from low- and middle-income countries, highlighting the need for standardized definitions and multicenter research.

INTRODUCTION

Acute kidney injury (AKI) is characterized by a sudden and often reversible decline in renal function. It is typically classified into prerenal, intrarenal, and postrenal categories based on the underlying pathophysiological mechanisms. The prevalence of AKI varies significantly, ranging from less than 1% to 66% across different studies, contingent upon the patient cohort, geographical location of the study, and the diagnostic criteria employed to define AKI[1].

AKI affects 40% to 80% of acute liver failure (ALF) patients[2]. Among cirrhosis patients, its prevalence can vary from 20% to 50%[3,4]. Irrespective of the cause, the onset of renal injury in patients with liver disease heralds increased morbidity and worsened prognosis. Consequently, understanding the pathophysiological mechanisms and implementing early interventions are crucial for improving outcomes.

Multiple factors predispose liver failure patients to renal injury. The association between advanced cirrhosis and hepatorenal syndrome (HRS-1 or HRS-AKI) underscores a distinct form of AKI predominantly driven by systemic and renal hemodynamic disturbances[4]. In contrast to chronic liver failure, the development of AKI in patients with ALF is far more complex, multifaceted, and distinct in mechanisms[5]. The key mechanisms leading to AKI in ALF include pre-renal insults, such as poor oral intake, use of laxatives, and aggressive decongestant therapy for raised intracranial pressure; ischemic acute tubular injury resulting from hemorrhagic or septic shock, congestive nephropathy, raised intra-abdominal pressure, or prolonged pre-renal insult; toxic acute tubular injury due to bile cast nephropathy or nephrotoxic drugs; and direct renal injury from the same insult that caused liver failure, for example, paracetamol poisoning.

Renal injury in patients with liver disease has been well-documented in the medical literature for the past seven to eight decades. Still, the primary focus of these studies has typically been AKI in patients with chronic liver disease, with very few studies evaluating its pathogenesis and treatment strategies in ALF[6,7]. Without studies assessing the contributions of different factors to renal injury in this patient population, implementing effective prophylactic measures and treatment strategies is challenging.

This systematic review aims to comprehensively assess the prevalence of AKI in patients with ALF, identify associated risk factors, evaluate its impact on clinical outcomes, and highlight key gaps in the existing research. Systematic reviews are a valuable tool to synthesize and summarize available data in a given field, providing a foundation for evidence-based practice and guiding future research[8].

The primary objective of this systematic review was to consolidate the epidemiological evidence on AKI in the context of ALF, with a focus on estimating its prevalence, identifying the risk factors contributing to its development, the associated clinical characteristics and outcomes, and highlighting gaps in the existing literature and reporting practices.

MATERIALS AND METHODS

Search strategy

Two investigators (Ahmed A and Gautam M) independently conducted a comprehensive literature search from January 2000 to December 2024 using multiple databases, including PubMed, MEDLINE, EMBASE, and Google Scholar, to identify relevant studies on AKI in ALF patients. We performed search using specific keywords phrase “acute kidney injury in acute liver failure”, “acute kidney failure in acute liver failure”, “acute renal insufficiency in acute liver failure”, “acute kidney injury in acute hepatic failure”, “acute kidney failure in acute hepatic failure”, “acute renal insufficiency in acute hepatic failure”. Moreover, we manually searched the references cited in the relevant articles. This approach aimed to capture the breadth of literature addressing the interplay between AKI and ALF, including studies with varying methodologies, geographical contexts, and clinical insights. The titles and abstracts of all documents retrieved through the literature search were thoroughly screened to assess their relevance. Studies found irrelevant to the research objectives and duplicate records were excluded from further analysis to ensure the final selection comprised unique and pertinent manuscripts. This multi-step approach was designed to maximize the comprehensiveness and rigor of the systematic review process. The literature search results are outlined in the PRISMA flow chart, as shown in Figure 1.

Figure 1  PRISMA flow chart.

Study selection

We established explicit inclusion and exclusion criteria for selecting studies. The literature was screened for eligibility based on these predefined criteria.

The inclusion criteria: (1) Original and full-length articles that reported epidemiological data on AKI in patients with ALF in the adult population; and (2) Articles published in the English language to ensure accurate interpretation and consistency during the review process.

The exclusion criteria: (1) Review articles; (2) Studies on AKI in acute-on-chronic liver failure; (3) Studies on AKI in chronic liver failure or cirrhosis; (4) Studies on chronic kidney disease in liver failure; (5) Paediatric studies; (6) Case reports; (7) Non-epidemiological investigations; and (8) Studies that failed to provide the required information. Any discordance in the selection of studies was solved using consensus, predefined inclusion, and exclusion criteria.

Data extraction

Two investigators (Ahmed A and Gautam M) independently screened the literature, resolving disagreements through discussion with other authors. Data was numerically extracted from each article that qualified the inclusion criteria and then presented in a Microsoft Excel spreadsheet (Windows 10 edition; Microsoft Corporation, Lisbon, Portugal). The extracted information was divided into different categories. The baseline information included the study title, study objective, design, setting, duration, and data regarding the inclusion and exclusion criteria, length of data collection, year of publication, and department in which the study was conducted. The demographic description of study participants was noted along with etiologies of ALF and AKI, stage of AKI, timeframe of AKI assessment, and risk factors of AKI in univariate and multivariate analysis. Outcomes were documented, focusing on the need for renal replacement therapy, mortality rates, liver transplantation, and transplant-free survival.

AKI was defined as a sudden decline in kidney function occurring within hours to days, typically within 48 hours. It is diagnosed using Risk, Injury, Failure, Loss of kidney function, and End-stage kidney disease (RIFLE) and Acute Kidney Injury Network (AKIN) criteria, but most recently, the Kidney Disease Improving Global Outcomes (KDIGO) guidelines are used as a standardized definition and staging system (Stages 1–3)[9]. Persistent AKI was defined as AKI that persisted for more than 48 hours based on the KDIGO criteria, as outlined in the consensus report of the Acute Disease Quality Initiative (ADQI) 16 Workgroup[10]. Transient AKI was defined as AKI lasting less than 48 hours[10].

ALF was defined as a rapid loss of liver function characterized by severe hepatocellular injury, leading to coagulopathy (international normalized ratio ≥ 1.5) and hepatic encephalopathy in patients without preexisting liver disease. ALF is categorized into hyperacute (< 7 days), acute (8 to 28 days), and subacute (29 days to 3 months) forms based on the time from jaundice to hepatic encephalopathy interval[11].

Methodological quality and outcome assessment

The methodological quality of eligible studies was assessed to evaluate the potential risk of bias in the retrieved information. We developed a customized bias assessment tool for this review to ensure a tailored approach. This questionnaire was designed to comprehensively evaluate study quality based on critical attributes relevant to the research objectives. The quality assessment tool consisted of a 5-point questionnaire assessing the following criteria: (1) Study design and setting: Study design and location were appropriately described; (2) Patient recruitment and description: The strategy is well-defined and suitable for the study objectives. The results report the number of patients recruited, and demographic and clinical characteristics of the recruited patients were described; (3) Operational definitions: Standard guidelines define AKI and ALF; (4) Outcome variables: Provide clear information about the outcome regarding the number of patients requiring renal replacement therapy, liver transplant free survival, and hospital mortality; and (5) Limitations: The discussion clearly defined the study's limitations. Studies that scored four or above were classified as high-quality evidence. Studies scoring three were considered moderate-quality evidence, while those scoring < 2 were considered poor-quality. This scoring system provided a standardized approach to assess and categorize the quality of studies included in the review.

Statistical analysis

Categorical data were expressed as counts and percentages. Information on the etiologies of AKI in ALF and hepatic and renal outcomes was extracted from individual studies and summarized using measures such as the mean, median, and range. Statistical analyses were conducted using GraphPad Prism version 5.0 (Carlsbad, CA, United States) and MedCalc software.

RESULTS

Our search identified a total of 1134 records. After screening and excluding articles that did not meet the study inclusion criteria, the full texts of 10 articles were assessed for eligibility. One study was excluded due to a language barrier, as shown in Figure 1. Ultimately, nine studies were included in the final analysis.

Description of studies

The methodological quality assessment revealed deficiencies in several aspects described in Table 1[12-20]. Four studies were rated as moderate quality, each scoring 3, while five studies were classified as high quality, with scores of 4 (2 studies) or 5 (3 studies). The systematic review included research articles published between January 2009 and December 2024. Studies were predominantly conducted by research groups in the United States and the United Kingdom[13-15,16,18]. Notably, research in this field has shown significant growth in recent years, with three articles published between 2022 and 2024 by research groups from various countries[18-20]. A total of nine studies were published over the last 15 years.

Table 1 Study quality assessment.

Ref.	Study design and setting are clearly defined	Eligibility and recruitment criteria of patients are clearly described	Operational definitions of ALF and AKI as per standard guidelines	Outcome variables described	Limitations of the study are clearly described	Score
Leithead et al[12]	Yes	No	Yes	No	Yes	3
O’Riordan et al[13]	Yes	Yes	No	Yes	Yes	4
Tujios et al[14]	Yes	Yes	Yes	Yes	Yes	4
Urrunaga et al[15]	Yes	No	No	Yes	Yes	3
Coelho et al[16]	Yes	Yes	Yes	Yes	Yes	5
Hadem et al[17]	No	Yes	No	Yes	Yes	3
Cardoso et al[18]	Yes	Yes	Yes	Yes	Yes	5
Imoto et al[19]	Yes	No	Yes	Yes	No	3
Fidalgo et al[20]	Yes	Yes	Yes	Yes	Yes	5

AKI: Acute kidney injury.

Out of the nine studies, eight were retrospective and one was a prospective study, with a total of 6031 participants. These studies comprised six single-center studies, two registry-based studies, and one multicenter study, as given in Table 2. Eight studies investigated both acetaminophen-related and non-acetaminophen-related ALF, while one study focused exclusively on acetaminophen-related ALF.

Table 2 Study characteristics.

Ref.	Home country	Study design	Length of data collection (years)	Number of centers	AKI criteria	Time frame of AKI assessment	Inclusion criteria	Exclusion criteria	Study location
Leithead et al[12]	United Kingdom	R	14	1	RIFLE	Hospital stay	ALF	MV, already on RRT	LTU
O’Riordan et al[13]	United Kingdom	R	8	1	AKIN	Hospital stay	Patients admitted to LITU with paracetamol poisoning	Liver transplant patients	LTU
Tujios et al[14]	United States	R	12	23	AKIN	Hospital stay	INR > 1.5, HE < 26 weeks of illness without liver disease	NA	NA
Urrunaga et al[15]	United States	R	10	Registry	Others	During the waitlisting period (median duration 3 days)	> 18 years ALF with status 1 or 1A for ALF transplant	NA	NA
Coelho et al[16]	Portugal	R	20	1	KDIGO	NA	18 years with ALF	CLD, CKD on RRT, RRT need before ICU	Gastroenterology, liver ICU
Hadem et al[17]	Germany	R	18	1	AKIN	Hospital stay	Hepatic encephalopathy with available renal function data	Overt sepsis	ICU
Cardoso et al[18]	United States	R	18	Registry	KDIGO	On day 1 of enrolment	> 18 years and ALF	< 3 days, follow up without renal function test on day 3, cirrhosis	NA
Imoto et al[19]	Japan	R	12	1	KDIGO	NA	ALF admitted to hospital	Malignancy, HIV	NA
Fidalgo et al[20]	Portugal	P	7	1	KDIGO	7 days of ICU stay	ALF/ALI > 16 years	NA	ICU

RIFLE: Risk, Injury, Failure, Loss of kidney function, and End-stage kidney disease; AKIN: Acute Kidney Injury Network; KDIGO: Kidney Disease Improving Global Outcomes; ALF: Acute liver failure; ALI: Acute liver injury; AKI: Acute kidney injury; CLD: Chronic liver disease; CKD: Chronic kidney disease; RRT: Renal replacement therapy; NA: Not available; LTU: Liver transplant unit; ICU: Intensive care unit; R: Retrospective; P: Prospective; INR: International normalized ratio; HE: Hepatic encephalopathy; MV: Mechanical ventilation.

Description of the study population

The sample size in these studies varied significantly, ranging from 51 to 2280 participants, as shown in Table 3. The mean age of the study populations was 41.86 ± 4.3 years with 67% female participants (range: 45.4% to 76.1%).

Table 3 Characteristics of study participants in studies describing acute kidney injury in acute liver failure patients, n (%).

Ref.	Total number of patients	Female gender	Age in years (mean)	Co-morbidities	Severity score	Number of patients developing AKI	AKI with ALF patients requiring RRT	Mortality (%)
Leithead et al[12]	308	175 (56.8)	39.7	NA	NA	208 (67.5)	145 (70)	112 (26.9)
O’Riordan et al[13]	302	178 (59)	37	NA	SOFA, MELD, APACHE II, KCH	239 (79.1)	182 (76)	74 (25)
Tujios et al[14]	1604	1106 (68.9)	NA	DM (n = 84; 5.2%), HTN (n = 93; 5.7%)	NA	1122 (69.9)	491 (43.7)	478 (29.8)
Urrunaga et al[15]	2280	1596 (70)	NA	NA	MELD	1271 (55.7)	NA	NA
Coelho et al[16]	51	28 (54.9)	NA	DM (n = 1; 5.9%), HTN (n = 8; 15.6%), CAD (n = 6; 9.8%), others (n = 5; 9.8%)	SOFA	34 (66.7)	5 (14.7)	22 (43.1)
Hadem et al[17]	134	102 (76.1)	42.5	NA	SOFA, SAPS III, KCH, BiLE	54 (40.3)	52 (96.2)	45 (33.6)
Cardoso et al[18]	1071	731 (68.3)	39	NA	MELD	744 (69.4)	221 (29.7)	324 (33.5)
Imoto et al[19]	174	79 (45.4)	49	NA	MELD	66 (37.9)	NA	20 (11.5)
Fidalgo et al[20]	104	44 (57.7)	44	NA	SOFA	57 (54.80)	32 (56.1)	32 (30.8)

DM: Diabetes mellitus; HT: Hypertension; CAD: Coronary artery disease; SOFA: Sequential Organ Failure Assessment; MELD: Model for End-Stage Liver Disease; APACHE: Acute Physiology and Chronic Health Evaluation; KCH: King's College Hospital criteria; HTN: Hypertension; SAPS: Simplified Acute Physiology Score; BiLE: Bilirubin Lactate and Etiology score; NA: Not available.

Data on co-morbidities were inconsistently reported across the studies. Diabetes mellitus was noted in up to 5.2% of patients in larger cohorts, while hypertension reached a prevalence of 15.6%. Coronary artery disease was less frequently observed. Notably, some studies lacked detailed information on co-morbid conditions.

The studies employed a variety of scoring systems to assess illness severity. The Sequential Organ Failure Assessment score was the most frequently used score in multiple studies. Other commonly applied tools included the Model for End-Stage Liver Disease score and the Acute Physiology and Chronic Health Evaluation score. Additionally, specific criteria like the King’s College Hospital criteria, Simplified Acute Physiology Score, and the Bilirubin Lactate and Etiology score were used in selected studies.

The studies had substantial heterogeneity in participant characteristics, co-morbidities, and AKI incidence, as shown in Table 3. The variability in retrieved data highlights the diverse methodologies employed in ALF-related AKI research, underscoring the need for standardized criteria to enhance comparability and facilitate clinical interpretation.

Table 4 summarizes the etiologies of ALF reported in different studies on AKI in ALF. Except for the Japanese cohort (Imoto et al[19]), drug-induced ALF was the predominant cause in most series, accounting for approximately one-quarter to all cases. Among drug-related etiologies, paracetamol was the most frequent offending agent, responsible for 100% of cases in O’Riordan et al[13] and Tujios et al[14], and for more than 80% of cases in Leithead et al[12] and Urrunaga et al[15] (86.7%).

Table 4 Etiologies of acute liver failure in different studies on acute kidney injury in acute liver failure, n (%).

Ref.	Drugs	Viral hepatitis	AIH	Budd-Chiari syndrome	Pregnancy	Wilson’s disease	Shock/ischemia	Other/undetermined
Leithead et al[12]	242 (78.5). PCM (89.6), non PCM (10.3)	46 (14.9)	5 (1.6)	5 (1.6)	3 (0.9)	2 (0.6)	0 (0)	5 (1.6)
O’Riordan et al[13]	302 (100). PCM (100)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)
Tujios et al[14]	738 (46). PCM (100)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	84 (5.2)	783 (48.8)
Urrunaga et al[15]	777 (34). PCM (86.7), non PCM (13.2)	216 (9.4)	118 (5.1)	3 (0.1)	19 (0.8)	13 (0.6)	0 (0)	1134 (49.7)
Coelho et al[16]	12 (23.5)	15 (29.4)	3 (5.8)	0 (0)	2 (3.9)	0 (0)	8 (15.6)	11 (21.6)
Hadem et al[17]	45 (33.8). PCM (28.9), non PCM (71.1)	31 (23.3)	2 (1.5)	9 (6.7)	0 (0)	3 (2.2)	1 (0.7)	43 (32.3)
Cardoso et al[18]	497 (46.4)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	0 (0)	573 (53.5)
Imoto et al[19]	5 (2.8). PCM (0), non PCM (100)	71 (40.8)	17 (9.8)	0 (0)	0 (0)	0 (0)	0 (0)	81 (46.5)
Fidalgo et al[20]	36 (34.6). PCM (50), non PCM (50)	11 (10.5)	0 (0)	0 (0)	0 (0)	0 (0)	8 (7.6)	49 (47.1)

PCM: Paracetamol; AIH: Autoimmune hepatitis.

Viral hepatitis was the second most common cause overall, accounting for 40.8% of cases in the Japanese study done by Imoto et al[19], 29.4% in Coelho et al[16], and 9.4% in Urrunaga et al[15]. Autoimmune hepatitis was reported in ≤ 9.8% of cases, Budd–Chiari syndrome in ≤ 6.7%, and pregnancy-related ALF in ≤ 3.9%. Wilson’s disease was rare (≤ 2.2%). “Other” or undetermined causes accounted for a substantial proportion of cases in several studies, including 48.8% in Tujios et al[14], 53.5% in Cardoso et al[18], and 47.1% in Fidalgo et al[20].

Incidence of AKI in ALF patients

A total of 6031 ALF patients were studied in nine studies, out of which 3795 (62.9%) patients developed AKI. The percentage of patients developing AKI varied widely, from 37.9% Imoto et al[19], to 79.1% O’Riordan et al[13]. Among the seven studies explicitly reporting AKI staging, the need for renal replacement therapy ranged from 14.7% to 96.2%. None of the studies described the cause or mechanism of AKI in these patients.

Most studies did not differentiate between transient and persistent AKI. However, two studies, Coelho et al[16], and Cardoso et al[18], reported this distinction. In the study conducted by Coelho et al[16], 29.4% patients had transient AKI while 70.5% had persistent AKI. In the study done by Cardoso et al[18], 9.6% had transient AKI, 80.7% had persistent AKI, and 9.5% experienced late-onset AKI (AKI after one day of admission).

The study done by Imoto et al[19], has used only coagulopathy in patients with deranged liver function as the diagnostic criteria of ALF, and they have further divided the patients into ALF with hepatic coma and ALF without hepatic coma. Development of AKI was more common in patients with hepatic coma as compared to patients without hepatic coma, 66.0% vs 38.5% respectively.

Risk factors for AKI in ALF patients

Only three studies have described the risk factors for developing AKI among ALF patients. Additionally, there is significant variability in the risk factors evaluated by these studies.

The study done by Leithead et al[12], in United Kingdom found systemic inflammatory response syndrome [odds ratio (OR) = 2.42 (95%CI: 1.17-5.00)], paracetamol poisoning [OR = 10.72 (95%CI: 4.24-27.09)], shock [OR = 7.01 (95%CI: 3.06-16.07)], King's College Hospital criteria [OR = 6.33 (95%CI: 2.65-15.13)] and infection [OR = 1.93 (95%CI: 0.93-4.02)] as independent risk factors for AKI in multivariate analysis.

Another study from United Kingdom done on paracetamol poisoning patients reported need for mechanical ventilation [OR = 4.78 (95%CI: 4.19-5.45)], vasopressor support [OR = 4.95 (95%CI: 1.02-24.81)], high admission phosphate [OR = 15.18 (95%CI: 3.46-66.02)], admission sodium [OR = 0.86 (95%CI: 0.77-0.96)], day-3 haematocrit [OR = 0.86 (95%CI: 0.77-0.96), and high day-3 Lactate [OR = 3.74 (95%CI: 1.77-7.92)] as independent risk factors for development of AKI in patients with paracetamol induced ALF[13].

Additionally, a study from Japan including 174 ALF patients reported TNF-α [OR = 1.647 (95%CI: 1.012–2.682)] as an independent risk factor linked with AKI (P = 0.0103)[19].

Mortality

Across the studies included, the average mortality rate was 29.3%, ranging from 11.5% to 43.1%. Notably, transplant-free survival was consistently higher among patients without AKI compared to those with AKI, as shown in Figure 2.

Figure 2 Comparison of liver transplant free survival (%) in non-acute kidney injury vs acute kidney injury. AKI: Acute kidney injury.

Among the various risk factors independently associated with mortality in ALF patients, kidney injury emerged as a significant predictor in multiple studies. Leithead et al[12], reported that AKI was associated with increased mortality at hospital discharge, with a hazard ratio (HR) of 5.48 (95%CI: 2.0–13.64) in multivariate analysis. Similarly, Fidalgo et al[20], found AKI to be an independent risk factor for all-cause mortality within the first seven days of ICU stay, with an HR of 11.61 (95%CI: 1.49–90.34). Urrunaga et al[15], identified stage III AKI as a predictor of mortality within the first six weeks in patients waitlisted for liver transplantation, with an HR of 2.77 (95%CI: 2.071–3.705). Hadem et al[17], also reported stage III AKI to be significantly associated with 28-day mortality, with an HR of 3.48 (95%CI: 1.75–6.93).

DISCUSSION

Our study highlights a high rate of AKI (62.9%) among patients with ALF, a complication that often precedes a more prolonged and complex hospital course. This is the first systematic review to synthesise the available literature on AKI in ALF. The findings should be interpreted in light of substantial heterogeneity across studies, including differences in study design (retrospective vs prospective), patient populations (acetaminophen vs non-acetaminophen ALF), diagnostic criteria for AKI (RIFLE, AKIN, KDIGO), and ALF (O’Grady vs Japanese classification). Furthermore, the length of the data collection ranges from 7 to 20 years, with at least three studies having a data collection period of more than 15 years. During this duration, changes in risk factors, diagnostic criteria, supportive care, and transplant practices potentially limit direct comparability.

A unified framework for identifying risk factors is also lacking, representing a key research gap. Several studies had missing or incomplete critical data, and none reported the underlying etiology of AKI (e.g., hypoperfusion, nephrotoxicity). Moreover, the pathophysiological mechanisms were not explored in any study, underscoring the need for future research in this area. Knowing the mechanism is important to guide the preventive measures, therapeutic interventions, and improve clinical outcomes, for example, ischemic AKI has been shown to have a better prognosis than septic AKI[21].

In contrast to ALF, cirrhosis, and chronic liver disease-related AKI are well studied in the literature. Lekakis et al[22] conducted a meta-analysis of 5202232 cirrhotic patients in 73 studies. The authors reported that AKI was common in hospitalized cirrhotic patients (29%) as well as stable outpatients (28%)[22]. On admission, AKI was most frequent in patients with infection or sepsis/septic shock (47%), followed by those with hepatic encephalopathy (41%). Disease severity significantly influenced AKI risk, with ICU patients experiencing the highest incidence (61%). In 2023, the International Club of Ascites and the ADQI convened to refine diagnostic criteria, outline graded recommendations for evaluation and management, and define research priorities in cirrhosis-related AKI[23]. In contrast, there remains a notable lack of studies and clinical guidelines addressing AKI in the context of ALF within the medical literature.

Our study also underscores the critical lack of data regarding AKI in ALF from low- and middle-income countries (LMICs). The impact of this bias on pooled estimates could not be quantified due to the lack of comparable LMIC data; however, it is plausible that variations in disease etiology, healthcare resources, and management practices in LMICs may influence AKI incidence and outcomes. Future multicenter studies from diverse geographic and economic settings are warranted to address this gap.

Recognizing geographical variations in the etiology of ALF is essential, as these differences influence the underlying pathophysiology and the approach to AKI management[24,25]. In the Asia-Pacific region, viral hepatitis remains the most common cause of ALF, whereas paracetamol (acetaminophen) overdose is the leading cause in Europe and North America. Even when drug-induced ALF occurs in the Asia-Pacific region, it is more often linked to anti-tubercular medications or herbal remedies rather than paracetamol[26,27].

The frequency and nature of AKI in ALF patients may vary depending on the underlying cause of liver failure and remain a subject insufficiently explored in the literature[28].

A notable example is obstetric ALF, a distinct and understudied clinical entity. In a study by Casey et al[29] conducted in the United States, the clinical features and outcomes of seventy pregnant patients who developed ALF were analyzed. The etiologies were categorized into pregnancy-associated acute liver disease (PAALD), acetaminophen poisoning (APAP), and other causes. Notably, admission serum creatinine levels were highest in the PAALD group (median 2.4 mg/dL, range 0.4–5.8 mg/dL), followed by the APAP group (median 1.4 mg/dL, range 0.3–5.4 mg/dL), and the 'other' group (median 0.7 mg/dL, range 0.3–2.7 mg/dL).

These findings suggest that PAALD-related ALF may be more frequently associated with kidney injury than other etiologies of ALF. One of the reasons for this could be a shared pathogenic pathway involving endothelial dysfunction and microcirculatory disturbances caused by elevated soluble fms-like tyrosine kinase-1 and soluble endoglin levels[30]. Recognizing this link is crucial for guiding timely pregnancy termination and initiating early, aggressive renal monitoring and nephroprotective strategies in pregnant patients with hepatic dysfunction.

As with many diseases, early identification of the pathological process enables timely intervention and improves outcomes. Biomarkers play a crucial role in early disease identification in its subclinical stage. In this regard, a noteworthy study by Imoto et al[19], conducted in Japan, examined the role of proinflammatory cytokines in developing AKI among ALF patients. The study found significantly higher levels of TNF-α in the AKI group compared to the non-AKI group (175.1 ± 227.9 pg/mL vs 52.9 ± 56.6 pg/mL; P = 0.0039). Incorporating such markers into diagnostic algorithms and, in the future, AI-driven predictive models could enable targeted, timely interventions, particularly when tailored to specific ALF subtypes[31].

Limitations

The lack of data on specific causes and risk factors for AKI in this population constrains our findings. High heterogeneity, limited data on etiology-specific AKI, and the retrospective design of most studies were the key limitations of our study. The heterogeneity in reporting methods across various studies posed challenges for direct comparisons. Most available data came from Europe and the USA, leaving other geographical regions underrepresented. Given potential regional differences in the causes and risk factors for AKI in ALF, further research is needed to address these gaps.

In light of the high incidence and substantial clinical impact of AKI in ALF, there is an urgent need for standardized diagnostic criteria, uniform reporting practices, and targeted interventional studies to guide clinical decision-making.

CONCLUSION

AKI is a common and serious complication in patients with ALF, with a prevalence ranging from 25% to 70% across studies. The development of AKI in this population is associated with increased morbidity and mortality. Variability in definitions, diagnostic criteria, and reporting practices poses challenges in accurately assessing incidence and outcomes. This review also reveals a marked lack of data from LMICs, where differences in etiology, healthcare resources, and management practices may significantly influence AKI patterns. These gaps underscore the need for standardized AKI definitions in ALF, early recognition, and targeted management strategies. Future multicentre prospective studies should aim to define risk factors better, refine diagnostic tools, and develop effective preventive and therapeutic interventions.