Revised: February 7, 2026
Accepted: March 24, 2026
Published online: September 25, 2026
Processing time: 194 Days and 20.3 Hours
Acute kidney injury (AKI) complicates up to 65% of hospitalizations for decom
To evaluate the diagnostic and prognostic utility of NGAL, cystatin C, and RRI for early AKI detection, subtyping [pre-renal, hepatorenal syndrome (HRS), acute tubular necrosis (ATN)], and mortality risk stratification in cirrhotic patients.
In this prospective cohort study, 120 cirrhotic patients with ascites were enrolled and categorized into three groups: Non-AKI (n = 42), pre-renal AKI (n = 43), and intrinsic AKI (n = 35), the latter subdivided into HRS (n = 20) and ATN (n = 15). Serum NGAL and cystatin C were quantified using commercially available en
AKI was present in 65% (78/120) of patients. Of these, 52 (66.7%) had AKI at admission, and 26 (33.3%) developed AKI during hospitalization. NGAL distinguished AKI with area under the curve (AUC) of 0.91 (95%CI: 0.87-0.95) [cutoff > 150 ng/mL, sensitivity of 90% (85%-94%), specificity of 85% (78%-90%)]. RRI > 0.74 identified HRS with AUC of 0.89 (95%CI: 0.83-0.94) with sensitivity of 82%, specificity of 88%. Cystatin C showed moderate diagnostic value [AUC of 0.86 (95%CI: 0.81-0.91)]. Urinary granular casts were specific for ATN (100%), while renin and aldo
NGAL and RRI outperform creatinine in early AKI detection, subtyping, and prognostication in cirrhotic patients, while cystatin C adds diagnostic value, particularly in ambiguous cases. An NGAL-RRI-based algorithm, enhanced by cystatin C when needed, represents a practical, resource-efficient tool for AKI triage in resource-limited settings.
Core Tip: Acute kidney injury (AKI) in cirrhosis is frequently underdiagnosed in resource-limited settings due to the unreliability of serum creatinine. This prospective Egyptian study, conducted in the hepatitis C virus-endemic population, demonstrates that integrating serum neutrophil gelatinase-associated lipocalin, cystatin C, and renal resistive index enables earlier AKI detection, accurate subtyping (pre-renal, hepatorenal syndrome, acute tubular necrosis), and improved short-term prognostication. We provide the first region-specific evidence that a simple neutrophil gelatinase-associated lipocalin-renal resistive index-based algorithm, supported by point-of-care ultrasound, can guide clinical triage and reduce unnecessary intensive care unit utilization in real-world Egyptian practice. These findings offer a practical, scalable diagnostic framework for cirrhosis-associated AKI in low-income and middle-income countries.
- Citation: Othman AAA, Mohamed MM, Eladl MM, Elsayed FMA. Integrated biomarkers and renal Doppler for early acute kidney injury diagnosis in hepatitis C virus cirrhosis. World J Nephrol 2026; 15(3): 119581
- URL: https://www.wjgnet.com/2220-6124/full/v15/i3/119581.htm
- DOI: https://dx.doi.org/10.5527/wjn.119581
Acute kidney injury (AKI) complicates nearly 65% of hospitalizations for decompensated cirrhosis globally, with mortality rates exceeding 60% in severe cases[1]. In Egypt, where hepatitis C virus (HCV)-related cirrhosis remains endemic and metabolic liver diseases are rising, AKI poses a particularly dire threat due to late diagnosis and limited access to renal replacement therapy (RRT). HCV-related cirrhosis accounts for up to 60% of AKI cases, yet regional biomarker validation is lacking[2]. Traditional markers like serum creatinine fail to detect AKI early in cirrhotic patients, as reduced hepatic creatinine production and sarcopenia mask renal dysfunction until irreversible damage occurs[3]. This diagnostic delay is exacerbated in resource-constrained settings, where specialized tests like fractional excretion of sodium (FENa) are often impractical[4].
International guidelines endorse neutrophil gelatinase-associated lipocalin (NGAL) and cystatin C as superior alternatives to serum creatinine for early AKI detection, with NGAL indicating tubular injury and cystatin C providing muscle mass-independent glomerular filtration rate (GFR) estimation[5]. However, their combined utility with renal resistive index (RRI) remains unvalidated in HCV-endemic regions like Egypt, where cirrhosis epidemiology differs markedly from Western alcohol-driven cohorts[6]. This gap is critical, as accurate AKI subtyping is essential for tailored therapy in resource-limited settings. Moreover, while Doppler ultrasonography (RRI) shows potential for differentiating hepatorenal syndrome (HRS) from other AKI types, its utility in real-world Egyptian hospitals remains untested[7]. This gap is critical, as HRS management (e.g., vasoconstrictors) differs radically from acute tubular necrosis (ATN) (fluid resu
We hypothesized that a combined biomarker panel (NGAL, cystatin C) and RRI would outperform serum creatinine for early AKI detection and subtype differentiation in HCV-predominant cirrhosis. To test this, we conducted a pro
This prospective diagnostic accuracy study was conducted through a collaborative effort between the Hepatology and Nephrology Units at Zagazig University Hospitals, a tertiary referral center serving Egypt’s Nile Delta region. Between December 2023 and June 2024, we screened 187 consecutive patients presenting with decompensated cirrhosis and ultrasonography-confirmed ascites. of whom 120 (64.2%) met all inclusion criteria and were included in the final study cohort, with exclusions due to end-stage renal disease (n = 22), hepatocellular carcinoma (HCC) (n = 15), and late presentation (> 24 hours post-admission, n = 30). The final analyzed cohort comprised 120 participants (mean age 54.2 ± 11.3 years; 68% male), reflecting the demographic profile of cirrhosis in our region, where hepatitis C and metabolic dysfunction-associated steatotic liver disease predominate.
Participants were classified into three clinically distinct groups based on AKI status: (1) Patients without AKI (n = 42), serving as the reference group for biomarker baselines; (2) Those with pre-renal AKI (n = 43), identified by rapid creatinine improvement after volume expansion; and (3) Intrinsic AKI cases (n = 35), further subdivided into HRS (n = 20) meeting ICA-AKI criteria and ATN (n = 15) confirmed by granular casts on urinary microscopy. This group stratification was intentionally designed to address key clinical questions: (1) Whether NGAL and cystatin C could distinguish cirrhotic patients with vs without AKI; (2) If these biomarkers could differentiate reversible pre-renal states from intrinsic kidney injury; and (3) How reliably renal Doppler (RRI) could identify HRS specifically.
The sample size was determined through a priori power analysis using PASS 2023 software, based on prior epidemiological data from Egyptian cohorts showing 65% AKI prevalence in decompensated cirrhosis[2]. With α = 0.05 (two-tailed) and 90% power, 120 participants allowed detection of a minimum area under the curve (AUC) difference of 0.15 between novel biomarkers (NGAL, cystatin C) and conventional creatinine-based AKI diagnosis, accounting for a 15% attrition rate, which was deemed clinically significant based on prior studies validating NGAL and cystatin C in cirrhosis[5,6]. This calculation specifically powered the study for three critical comparisons: (1) NGAL vs creatinine for any AKI detection (primary endpoint); (2) RRI for HRS identification (secondary endpoint); and (3) The combined biomarker panel for AKI subtyping (exploratory endpoint). The assumed intraclass correlation of 0.3 between biomarkers was derived from pilot data at our institution involving 30 cirrhotic patients with serial measurements.
Ethical approval was obtained from the Institutional Review Board of Suez University (No. 9-SU-Med-2023). The approved study protocol identified Zagazig University Hospitals as the recruitment and data collection site. The study protocol was prospectively registered at the Pan African Clinical Trials Registry (PACTR202505555006910; registration date: April 18, 2025). The study strictly adhered to the ethical principles of the Declaration of Helsinki (2013 revision) and Egypt’s national regulations for biomedical research (Ministerial Decree 296/2021). Written informed consent was obtained from all participants after a detailed explanation in Arabic by trained hepatology fellows; for patients with hepatic encephalopathy, written consent was obtained from legally authorized representatives with subsequent patient re-consent upon cognitive improvement. Data confidentiality was maintained through pseudonymization using unique study IDs, with encrypted electronic records stored on password-protected hospital servers accessible only to the principal investigators. All laboratory procedures followed ISO 15189:2022 standards for medical laboratories, with periodic external quality assurance through the College of American Pathologists proficiency testing program.
The study population comprised adult patients with decompensated cirrhosis who met all inclusion criteria: (1) Confirmed diagnosis of cirrhosis (either histologically or through combined clinical, biochemical, and radiological evidence); (2) Presence of detectable ascites on abdominal ultrasonography; (3) Willingness to provide informed consent; and (4) Ability to complete the study procedures.
We excluded several patient groups to minimize confounding: (1) Pediatric patients (< 18 years) due to distinct AKI pathophysiology in developing kidneys; (2) Individuals with multifocal HCC as tumor-related inflammation may artificially elevate NGAL levels; (3) Patients with end-stage renal disease (estimated GFR < 15 mL/minute/1.73 m2 or on maintenance dialysis) given their fundamentally altered biomarker baselines; (4) Transplant recipients because immunosuppressive regimens affect tubular function; (5) Patients with sepsis or septic shock – defined per sepsis-3 criteria as infection with organ dysfunction [Sequential Organ Failure Assessment (SOFA) score ≥ 2] – as these conditions profoundly alter renal hemodynamics and biomarker profiles; (6) Patients with localized infections without systemic organ dysfunction (e.g., uncomplicated urinary tract infection, cellulitis) were included, as infection remains a clinically relevant predictor of AKI in cirrhosis; (7) Those with known thyroid disorders (to mitigate endocrine-driven elevations in cystatin C); (8) Those with recent upper gastrointestinal bleeding (< 2 weeks) since acute hemorrhage independently influences cystatin C concentrations; and (9) Patients presenting > 24 hours post-admission to ensure capture of early AKI biomarkers. These exclusion criteria were implemented through a two-stage screening process involving both emergency department physicians and hepatology consultants.
At enrollment, all participants underwent a comprehensive clinical evaluation performed by board-certified hepatologists. Baseline assessments included detailed medication reconciliation with particular attention to diuretic doses and nephrotoxic agents, thorough physical examination documenting ascites grade, peripheral edema, and signs of hepatic encephalopathy using West Haven criteria. Anthropometric measurements were standardized across participants: (1) Weight was measured using a calibrated bed scale (SECA 956) with subtraction of estimated ascites volume; (2) Height was determined by a wall-mounted stadiometer; and (3) Body mass index was calculated. Blood pressure measurements followed American Heart Association guidelines using automated cuffs (Omron HEM-7322) with appropriate sizing, taken after 10 minutes of rest in the supine position. Jugular venous pressure was assessed to estimate volume status, and all findings were recorded on standardized case report forms adapted from the European Association for the Study of Liver-Chronic Liver Failure consortium protocols.
To support the group stratification, all participants underwent standardized volume status assessment, including daily weight measurements, strict fluid balance documentation, and central venous pressure estimation when clinically indicated. Pre-renal AKI diagnosis required both biochemical evidence (FENa < 0.1%) and documented ≥ 20% creatinine improvement after 48 hours of controlled volume expansion with albumin-guided therapy, per ICA-AKI guidelines. HRS cases were adjudicated by a panel of three hepatologists using ICA criteria modified to exclude urinary sodium, as > 80% of our cirrhotic cohort was on chronic diuretic therapy at admission, rendering urinary sodium unreliable for HRS diagnosis. This adaptation aligns with real-world hepatology practice in high-diuretic-use settings, while ATN required both FENa > 0.2% and urinary granular casts on microscopic examination by two independent nephropathologists.
Volume status assessment was further refined using point-of-care ultrasound (POCUS) of the inferior vena cava (IVC) and lung fields. IVC collapsibility index (IVCCI) was measured during quiet respiration, with values < 40% suggestive of hypervolemia and > 60% indicating hypovolemia. Bilateral lung ultrasound was performed to detect B-lines (≥ 3 per field) suggestive of pulmonary congestion. These findings supported AKI phenotyping, especially in cases where clinical signs were equivocal. All prescriptions within the 7-day period preceding AKI onset (or hospital admission for non-AKI controls) were reviewed. Known nephrotoxins – including non-steroidal anti-inflammatory drugs, aminoglycosides, and proton pump inhibitors – were recorded and categorized by agent class, dose (standard vs high), and duration (≤ 3 days vs > 3 days). No formal washout period was applied due to the acute observational design; exposure was analyzed as binary (any/none) and as a multi-level categorical variable. Infection status was documented systematically using sepsis-3 criteria; cultures and inflammatory markers (C-reactive protein, procalcitonin when available) were recorded, and SOFA score were calculated at admission to distinguish localized infection from sepsis. Potential associations between nephrotoxic exposure and AKI incidence were explored via regression modeling.
Blood samples were collected at enrollment through venipuncture of antecubital veins using 21-gauge safety needles (BD Vacutainer) into serum separator tubes II advance (BD Vacutainer). Serum NGAL and cystatin C samples were collected within 24 hours of suspected AKI onset or hospital admission for non-AKI controls to capture early biomarker changes.
For NGAL and cystatin C measurements, samples were processed within 30 minutes of collection by centrifugation at 2500 rpm for 15 minutes at 4 °C (Eppendorf 5702R). Aliquots were stored at -80 °C (Thermo Scientific Forma 900 series) until batch analysis. Serum NGAL and cystatin C concentrations were determined using commercially available enzyme-linked immunosorbent assay (ELISA) kits (Sunred Biological Technology Co., Ltd, Shanghai, China), following the manufacturer’s protocols validated per CLSI EP15-A3 guidelines. The absorbance was measured using a microplate reader (BioTek ELx808, Agilent Technologies, Santa Clara, CA, United States). The intra- and inter-assay coefficients of variation were < 10%.
Routine biochemistry, including creatinine (enzymatic method), electrolytes, and liver function tests [including alanine aminotransferase, aspartate aminotransferase, alkaline phosphatase, gamma-glutamyl transferase, total and direct bilirubin, albumin, and prothrombin time/international normalised ratio (INR)], were measured upon admission using a Cobas 8000 analyzer (Roche Diagnostics, Basel, Switzerland) following standardized laboratory protocols.
In this study, POCUS denotes a focused, goal-directed renal and hemodynamic ultrasound assessment performed by trained faculty operators within a standardized clinical protocol, rather than a comprehensive diagnostic radiologic examination. RRI measurements were acquired using a standardized Doppler protocol on a GE Logiq E10 system with a C1-6-D transducer. After ensuring patients had fasted for ≥ 4 hours, examinations were conducted in the supine position with normal respiration. Three consecutive waveforms were obtained from interlobar arteries of both kidneys at a consistent 60° Doppler angle during end-expiration. All examinations were performed and interpreted by senior faculty members with advanced expertise in abdominal and renal ultrasonography, including professors of internal medicine and radiology, with all studies interpreted by two abdominal radiologists blinded to clinical data. Mean RRI values were calculated from six total measurements (three per kidney). Monthly quality control sessions, supervised by a senior radiologist, confirmed excellent inter-operator reliability (intra-class correlation coefficient > 0.85)[9], ensuring mea
For comprehensive AKI phenotyping, additional diagnostic tests were performed.
Urinary sediment analysis: This was performed in all enrolled patients (n = 120) to assess structural kidney injury. Fresh midstream urine samples were examined within 30 minutes of collection using phase-contrast microscopy (Olympus CX43, 400 × magnification). The presence of ≥ 5 granular casts per high-power field (hpf) in two consecutive samples was required for the diagnosis of ATN. Muddy brown casts were recorded when present and considered suggestive of tubular epithelial injury.
FENa: FENa was measured in all patients (n = 120) to aid in the etiologic classification of AKI. Paired serum and spot urine samples were collected within 15 minutes and analyzed on Cobas 8000 platforms. A FENa < 0.1%, after accounting for diuretic exposure, was used to define pre-renal AKI.
Renin-aldosterone axis evaluation: Renin and aldosterone levels were assessed in a targeted subgroup of 28 patients with clinically ambiguous features suggestive of HRS. Blood samples were obtained after 30 minutes of supine rest and processed via radioimmunoassay (DiaSorin Liaison XL). Results were interpreted against cirrhosis-adjusted reference ranges to support the diagnosis of HRS in diagnostically unclear cases.
The primary outcome was AKI incidence within 7 days, defined by AKIN criteria (absolute creatinine increase ≥ 0.3 mg/dL or ≥ 50% from baseline). Patients were enrolled at the point of clinical evaluation for AKI; thus, the cohort represents a pragmatic mix of AKI present on admission (prevalent AKI) and AKI developing during hospitalization (incident AKI). This design maximizes clinical applicability for evaluating biomarker utility in diagnosing and subtyping AKI at the time of clinical suspicion. AKI subtyping followed a rigorous adjudication protocol: (1) Pre-renal cases required both FENa < 0.1% and ≥ 20% creatinine improvement after volume expansion; (2) HRS was diagnosed using modified ICA-AKI criteria (excluding urinary sodium due to prevalent diuretic use); and (3) ATN demanded FENa > 0.2% with confirmatory urinary casts. Two independent hepatologists performed initial classification, with a third reviewer resolving dis
Secondary outcomes encompassed both therapeutic and prognostic measures: (1) Time-to-creatinine-recovery in pre-renal AKI (return to within 0.3 mg/dL of baseline); (2) Vasoconstrictor response in HRS (≥ 25% creatinine reduction after terlipressin); and (3) AKI etiology classification (pre-renal, HRS, or ATN) adjudicated by an independent committee using modified ICA criteria. Prognostic evaluation included 30-day mortality (verified through hospital records and family contact), renal recovery (creatinine within 0.3 mg/dL of baseline without RRT), and a composite endpoint of mortality, dialysis dependence, or intensive care unit (ICU) admission. All outcomes were assessed by investigators blinded to biomarker results until database lock.
AKI was staged according to updated ICA-KDIGO criteria: (1) Stage 1: Serum creatinine increase ≥ 0.3 mg/dL or 1.5-2-fold from baseline; (2) Stage 2: 2-3-fold increase; and (3) Stage 3: > 3-fold increase, creatinine > 4.0 mg/dL, or initiation of dialysis. This staging enabled evaluation of AKI severity and its correlation with outcomes. Follow-up was conducted via outpatient visits and telephone interviews.
Data were analyzed using SPSS version 28 (IBM Corp., Armonk, NY, United States) and R version 4.3.1 with packages pROC, caret, survival, and ggplot2 for receiver operating characteristic (ROC) analysis, XGBoost modeling, Cox reg
Diagnostic accuracy was evaluated using ROC curve analysis. AUC values were compared using DeLong’s method, and optimal diagnostic thresholds were derived from the cohort data by maximizing Youden’s index. ROC curves were generated using the pROC package in R. For exploratory biomarker-interaction analysis, an XGBoost machine-learning classifier was implemented using repeated nested cross-validation (CV) (5-fold inner loop, 10-fold outer loop, 10 repeats) to mitigate overfitting and provide robust internal performance estimates. Hyperparameter tuning was performed via Bayesian optimization within the inner loops. Model calibration was assessed visually using calibration curves and quantitatively with the Brier score and calibration slope; the Hosmer-Lemeshow test was not applied, given the non-parametric nature of tree-based ensembles.
Multivariable analyses were performed to account for confounding factors. Logistic regression models were used for binary outcomes, adjusted for Model for End-Stage Liver Disease includes serum sodium (MELD-Na) score, infection status, and diuretic use. Cox proportional hazards models were applied to time-to-event outcomes, with Schoenfeld residuals used to test proportionality assumptions. For analyses comparing AKI subtypes, additional adjustments were made for baseline NGAL levels and diuretic dosing.
Sensitivity analyses were conducted to assess the robustness of the findings. These included a comparison between complete-case analysis and multiple imputation to address minimal missing data (< 5%), as well as the exclusion of borderline FENa cases (0.1%-0.2%) to minimize diagnostic ambiguity, as values in this intermediate range often reflect diuretic exposure, altered tubular sodium handling in cirrhosis, or overlapping pathophysiologies that compromise clear phenotyping. This preserves phenotypic group integrity and enhances the internal validity of subtype comparisons. Additionally, the diagnostic performance of the key biomarkers (NGAL, RRI) was evaluated in sensitivity analyses stratified by AKI onset (prevalent vs incident). To explore factors associated with AKI development, a multivariate logistic regression model was constructed, including established risk variables: (1) MELD-Na; (2) Total bilirubin; (3) Serum albumin; (4) Infection status; (5) Diuretic use; and (6) Exposure to nephrotoxic medications modeled as a binary (any/none) and categorical (by drug class and duration) variable. No formal washout period was applied, given the acute observational design. Collinearity was excluded by calculating variance inflation factors.
Kaplan-Meier survival analysis was employed to assess 30-day mortality across AKI stages. Survival differences were evaluated using the log-rank test, while Cox models incorporated AKI stage as a time-dependent covariate, adjusted for MELD-Na, infection status, and baseline NGAL levels. All statistical tests were two-tailed, with a significance threshold set at P < 0.05. Multiplicity adjustment was not applied due to the exploratory nature of this diagnostic performance study. All study figures were generated using R version 4.3.1 with ggplot2 and related visualization packages. Correction for multiple comparisons was applied, as the primary and secondary analyses were predefined and hypothesis-driven.
Among the 120 patients with decompensated cirrhosis enrolled in this study, 78 (65%) met criteria for AKI at the point of clinical evaluation, while 42 (35%) did not develop AKI and served as biomarker baseline controls (Table 1). The AKI cohort comprised a pragmatic mix of presentations: (1) 52 patients (66.7% of AKI cases) had AKI present at hospital admission (prevalent AKI); and (2) 26 patients (33.3%) developed AKI after admission (incident AKI). Within the AKI cohort, 43 cases (55.1%) were classified as pre-renal AKI based on response to volume expansion and FENa < 0.1%. Intrinsic AKI accounted for 35 cases (44.9%), including 20 patients (25.6%) with HRS meeting ICA-AKI criteria, and 15 patients (19.2%) diagnosed with ATN by microscopy-confirmed granular casts.
| Variable | Non-AKI (n = 42) | Pre-renal AKI (n = 43) | Hepatorenal syndrome | Acute tubular necrosis (n = 15) | P value |
| Age (years) | 52.4 ± 11.0 | 54.1 ± 10.8 | 56.6 ± 9.7 | 55.1 ± 10.4 | 0.12 |
| Male sex | 24 (57.1) | 31 (72.1) | 14 (70.0) | 11 (73.3) | 0.34 |
| Model for End-Stage Liver Disease includes serum sodium score | 16.8 ± 4.2 | 20.3 ± 5.1b | 24.2 ± 5.4 | 23.7 ± 5.3 | < 0.001 |
| Serum creatinine (mg/dL) | 0.9 ± 0.3 | 1.8 ± 0.4a | 2.0 ± 0.6 | 2.2 ± 0.7 | < 0.001 |
| Fractional excretion of sodium (%) | 0.08 ± 0.1 | 0.05 ± 0.1a | 0.11 ± 0.1c | 0.26 ± 0.2b,c | < 0.001 |
| Total bilirubin (mg/dL) | 2.3 ± 1.2 | 3.8 ± 1.9a | 5.7 ± 2.0 | 5.3 ± 2.2 | < 0.001 |
| Albumin (g/dL) | 3.2 ± 0.5 | 2.9 ± 0.4a | 2.6 ± 0.3 | 2.4 ± 0.4 | < 0.001 |
| International normalised ratio | 1.3 ± 0.2 | 1.5 ± 0.3a | 1.9 ± 0.4 | 1.8 ± 0.3 | < 0.001 |
| Alanine aminotransferase (U/L) | 32 ± 14 | 36 ± 16 | 40 ± 17 | 39 ± 15 | 0.09 |
| Aspartate aminotransferase (U/L) | 45 ± 18 | 49 ± 20 | 61 ± 22a | 58 ± 21a | 0.03 |
| AKI onset | - | 0.55 | |||
| Prevalent (at admission) | - | 30 (69.8) | 12 (60.0) | 10 (66.7) | |
| Incident (post-admission) | - | 13 (30.2) | 8 (40.0) | 5 (33.3) |
Demographic variables, including age and sex distribution, were comparable across all study groups (P > 0.05), minimizing the likelihood of confounding in biomarker comparisons. MELD-Na scores and serum creatinine levels were significantly elevated in HRS and ATN groups compared to pre-renal AKI and non-AKI controls (P < 0.001), un
Liver function tests revealed a progressive deterioration in hepatic synthetic and excretory capacity across AKI phenotypes. Total bilirubin levels were highest in HRS (5.7 ± 2.0 mg/dL) and ATN (5.3 ± 2.2 mg/dL). Albumin levels were significantly lower in these groups, indicating impaired protein synthesis. INR values followed a similar gradient, with prolonged coagulation times in HRS (1.9 ± 0.4) and ATN (1.8 ± 0.3).
We performed a sensitivity analysis to evaluate whether biomarker performance was consistent across the clinical spectrum of AKI presentation. The distribution of AKI subtypes (pre-renal, HRS, ATN) did not differ significantly between prevalent and incident AKI cases (P = 0.55; Table 1). Importantly, the diagnostic accuracy of serum NGAL for distinguishing any AKI from non-AKI remained excellent in both subgroups: (1) AUC of 0.92 (95%CI: 0.87-0.97) for prevalent AKI; and (2) AUC of 0.88 (95%CI: 0.80-0.96) for incident AKI. The biomarker profiles characteristic of each AKI subtype (e.g., elevated RRI in HRS, high NGAL in ATN) remained consistent regardless of AKI onset timing, supporting our primary combined cohort analysis for subtyping.
Evaluation of novel biomarkers revealed that NGAL and cystatin C were significantly superior to serum creatinine in detecting early AKI (Table 2 and Figure 1). Using Youden’s index to maximize sensitivity and specificity, the optimal data-driven cutoff for NGAL was 150 ng/mL. At a threshold of 150 ng/mL, serum NGAL distinguished AKI from non-AKI with an AUC of 0.91 (95%CI: 0.87-0.95), offering 90% sensitivity and 85% specificity. NGAL’s superior AUC reflects its early rise after tubular injury, enabling earlier detection than creatinine, highlighting its utility in mitigating diagnostic delays and enabling earlier intervention in cirrhosis-associated AKI. In clinical practice, this would allow for earlier intervention, potentially preventing further renal deterioration in cirrhotic patients. Cystatin C also demonstrated strong diagnostic performance [AUC of 0.86 (95%CI: 0.81-0.91)], offering a reliable creatinine-independent alternative, particularly beneficial in sarcopenic or cachectic patients, which indicates its reliability as an alternative marker, especially in cases where factors such as liver dysfunction may influence serum creatinine.
| Marker | Area under the curve (95%CI) | Cutoff | Sensitivity (%) (95%CI) | Specificity (%) (95%CI) |
| Neutrophil gelatinase-associated lipocalin | 0.91 (0.87-0.95) | > 150 ng/mL | 90 (85-94) | 85 (78-90) |
| Cystatin C | 0.86 (0.81-0.91) | > 1.5 mg/L | 84 (77-89) | 80 (72-86) |
| Renal resistive index | 0.89 (0.83-0.94) | > 0.74 | 82 (74-88) | 88 (81-93) |
| Creatinine | 0.73 (0.66-0.80) | > 1.2 mg/dL | 68 (60-76) | 62 (54-70) |
Importantly, RRI > 0.74 emerged as a strong diagnostic marker for HRS [AUC of 0.89 (95%CI: 0.83-0.94), sensitivity of 82%, specificity of 88%], supporting its value as a point-of-care Doppler tool in differentiating vasoconstrictive renal dysfunction from other causes of AKI. This result underscores the potential clinical utility of Doppler ultrasonography in identifying patients with HRS, a subgroup of AKI that typically has a poor prognosis and for whom targeted inter
The performance of serum creatinine [AUC of 0.73 (95%CI: 0.66-0.80)], though lower than the other biomarkers, still provides useful diagnostic information, particularly for routine monitoring. However, its lower sensitivity and specificity in early AKI detection may limit its utility in distinguishing AKI from other causes of renal impairment, especially in cirrhosis, where creatinine levels can be affected by factors other than kidney injury.
Specialized diagnostic markers supported the phenotypic stratification of intrinsic AKI (Table 3 and Figure 2). Granular casts ≥ 5 per hpf were present in 100% (15/15) of ATN cases and absent in all HRS (0/20) and pre-renal AKI (0/43) cases, affirming their 100% specificity and positive predictive value for tubular injury. This finding underscores the importance of urinary sediment analysis in distinguishing structural ATN from functional AKI subtypes in decompensated cirrhosis.
| Test | Pre-renal acute kidney injury (n = 43) | Hepatorenal syndrome (n = 20) | Acute tubular necrosis (n = 15) | P value |
| Granular casts ≥ 5/high-power field | 0% (0/43) | 0% (0/20) | 100% (15/15) | < 0.001 |
| Renin (ng/mL) | 5.8 ± 2.1 | 12.1 ± 4.0 | 6.3 ± 2.4 | 0.01 |
| Aldosterone (pg/mL) | 98 ± 32 | 240 ± 68 | 112 ± 44 | 0.03 |
Hormonal profiling revealed that HRS patients (n = 20) had significantly higher circulating renin (12.1 ± 4.0 ng/mL) and aldosterone (240 ± 68 pg/mL) levels compared to pre-renal AKI (n = 43) and ATN (n = 15) groups, consistent with systemic vasodysregulation in HRS. Exact values for renin (pre-renal: 5.8 ± 2.1 ng/mL; HRS: 12.1 ± 4.0 ng/mL; ATN: 6.3 ± 2.4 ng/mL) and aldosterone (pre-renal: 98 ± 32 pg/mL; HRS: 240 ± 68 pg/mL; ATN: 112 ± 44 pg/mL) are depicted in Figure 2. In contrast, the lower values in the ATN and pre-renal groups reflect different underlying mechanisms: (1) Volume depletion in pre-renal AKI; and (2) Tubular damage in ATN. This result is crucial for understanding the pathophysiology of HRS, where the kidneys’ compensatory mechanisms (e.g., renin-aldosterone system activation) are inappropriately upregulated in response to reduced renal perfusion, leading to further renal dysfunction. In contrast, the lower renin and aldosterone levels in the pre-renal and ATN groups reflect different underlying mechanisms, such as volume depletion in pre-renal AKI and tubular injury in ATN, where the compensatory hormonal response is less pronounced.
These findings suggest that a combination of urinary sediment analysis and hormonal profiling may provide valuable insights into the specific mechanisms driving AKI in cirrhosis, enhancing our ability to personalize treatment strategies.
Thirty-day outcomes were significantly associated with biomarker levels at admission (Table 4). Patients with NGAL > 150 ng/mL had a 3.1-fold increased risk of mortality [adjusted hazard ratio (HR) = 3.1, 95%CI: 1.7-5.5, P < 0.001], and those with RRI > 0.74 had a 2.2-fold higher risk of poor renal recovery or death. These findings underscore the prognostic value of NGAL and RRI for short-term outcomes in decompensated cirrhosis, providing valuable tools for early risk stratification. The independent association between elevated NGAL levels (> 150 ng/mL) and 30-day mortality is consistent with its known correlation with AKI severity and systemic inflammation, reinforcing its role as a strong prognostic indicator in this patient population.
| Variable | Adjusted hazard ratio (95%CI) | P value |
| Neutrophil gelatinase-associated lipocalin > 150 ng/mL | 3.1 (1.7-5.5) | < 0.001 |
| Renal resistive index > 0.74 | 2.2 (1.3-3.7) | 0.003 |
| Model for End-Stage Liver Disease includes serum sodium > 25 | 3.9 (2.1-7.2) | 0.01 |
Additionally, a MELD-Na score > 25 was independently associated with mortality (adjusted HR = 3.9, 95%CI: 2.1-7.2; P = 0.01), highlighting the complex interplay between hepatic and renal function in decompensated cirrhosis.
These results suggest that incorporating both biomarkers and established scoring systems like MELD-Na can significantly enhance prognostic accuracy in cirrhosis-associated AKI. Early identification of patients at high risk for mortality or poor renal recovery enables targeted management strategies aimed at improving outcomes (Figure 3).
In Cox models adjusted for MELD-Na, infection status, and baseline NGAL, AKI stage remained independently associated with 30-day mortality (Table 5). Compared to non-AKI patients, adjusted HRs were 2.8 (95%CI: 0.7-11.2, P = 0.14) for stage 1, 4.6 (95%CI: 1.5-14.3, P = 0.008) for stage 2, and 9.3 (95%CI: 3.4-25.4, P < 0.001) for stage 3. Stage 1 AKI was associated with a non-significant increase in mortality risk, whereas stages 2 and 3 showed statistically and clinically meaningful elevations in hazard.
| AKI stage | n | 30-day mortality count | 30-day mortality percentage | Cox P value | Comment | Hazard ratio (95%CI) |
| No AKI | 42 | 1 | 2.4% | - | Baseline comparator | 1.0 (reference) |
| Stage 1 | 28 | 2 | 7.1% | 0.14 | Mild injury, lower risk | 2.8 (0.7-10.5) |
| Stage 2 | 25 | 5 | 20.0% | 0.008 | Intermediate severity | 4.6 (1.5-13.8) |
| Stage 3 | 25 | 14 | 56.0% | < 0.001 | High-risk group, dialysis/hepatorenal syndrome | 9.3 (3.7-23.4) |
This analysis affirms that the stage of AKI significantly influences short-term mortality risk in cirrhosis patients. It also demonstrates that stage 3 AKI (n = 25) is associated with the highest mortality, which aligns with clinical observations that these patients often require dialysis or experience HRS. The visual gradient in mortality supports the stratification utility of the current AKI classification in cirrhosis care.
These findings support the clinical value of AKI staging, alongside biomarker profiling, in estimating short-term prognosis among patients with decompensated cirrhosis. The ability to stratify risk based on AKI stage enables clinicians to make more informed decisions regarding monitoring and management strategies, such as RRT and vasopressor support. Future models may enhance prognostic accuracy further by integrating AKI stage with dynamic changes in renal biomarkers and systemic inflammatory markers (Figure 4).
To explore interactions between biomarkers, we employed an XGBoost classification model, internally validated through repeated nested CV to mitigate overfitting. The NGAL-RRI combination achieved the highest cross-validated predictive performance, with a CV-AUC of 0.92 (95%CI: 0.89-0.95) (Table 6). The model also demonstrated good calibration (calibration slope of 0.96, Brier score of 0.12) and a balanced F1-score of 0.85. It significantly outperformed conventional models based on creatinine and cystatin C, demonstrating superior accuracy in AKI subtyping while maintaining robust internal validity.
| Model features | Area under the curve from repeated nested cross-validation (95%CI) | Accuracy | F1-score | Brier score | Calibration slope | Sensitivity (95%CI) | Specificity (95%CI) |
| Neutrophil gelatinase-associated lipocalin + RRI (XGBoost) | 0.92 (0.89-0.95) | 86% | 0.85 | 0.12 | 0.96 | 89% (83-93) | 84% (77%-89%) |
| Creatinine + cystatin C | 0.81 (0.75-0.86) | 74% | 0.73 | 0.18 | 0.91 | 76% (69-82) | 71% (63%-78%) |
| RRI alone | 0.78 (0.72-0.84) | 72% | 0.70 | 0.20 | 0.89 | 73% (66-80) | 69% (61%-76%) |
NGAL, a well-established urinary biomarker of tubular injury, in combination with RRI, a renal Doppler parameter reflecting renal vasculature and function, demonstrated superior diagnostic power. This finding suggests that the combination of urinary biomarkers and renal imaging tools can significantly improve the accuracy of AKI classification and risk stratification, potentially allowing for earlier and more precise diagnosis of intrinsic AKI in patients with decompensated cirrhosis.
Sensitivity analyses, which excluded ntermediate FENa values (0.1%-0.2%, n = 11), a range often ambiguous due to diuretic exposure or altered tubular sodium handling in cirrhosis, revealed further improvement in the cross-validated AUC for HRS, from 0.90 (0.86-0.94) to 0.92 (0.88-0.95), demonstrating the robustness of the XGBoost algorithm when applied to more clearly defined phenotypes. This demonstrates the robustness of the XGBoost algorithm when applied to more clearly defined phenotypes and highlights the potential of machine learning (ML) in refining diagnostic accuracy by minimizing the influence of ambiguous or borderline data. This adjustment enhances the clinical applicability of ML models in settings where certain variables may have ambiguous thresholds.
The results underline the importance of employing advanced analytics and ML techniques to identify subtle interactions between biomarkers that might not be apparent through traditional statistical methods. These findings could help refine diagnostic algorithms and support personalized medicine approaches for AKI management in decompensated cirrhosis.
To explore the theoretical impact of biomarker-guided triage, we performed a decision analysis simulating the imple
This decision model represents a retrospective, data-driven decision-support framework and was not prospectively implemented or evaluated for clinical outcomes.
Applying this triage model to our cohort was associated with a 28% reduction in potentially inappropriate ICU transfers (14 of 50 avoided), predominantly among patients with pre-renal or low-risk intrinsic AKI (Figure 5). Avoidance of ICU admission in these patients implies meaningful reductions in direct costs related to bed utilization, monitoring, and invasive interventions, as well as indirect benefits through improved ICU capacity allocation. Figure 6 provides a complementary stepped-care schematic illustrating the clinical translation of this decision tree into a sequential diagnostic and management pathway suitable for routine practice.
While cystatin C provided supplementary estimation of GFR, NGAL and RRI were prioritized within the model due to their superior performance in early AKI detection and hemodynamic characterization, respectively. Future refinements of this framework may explore the combined use of cystatin C and NGAL for patients with intermediate biomarker values (e.g., 100-200 ng/mL).
RRI findings were further validated by Doppler imaging, demonstrating its utility as a non-invasive diagnostic tool for differentiating between various forms of AKI in cirrhosis. As shown in Figure 7, representative Doppler waveforms clearly illustrate the differences in renal vasculature between the AKI subtypes.
In patients with HRS, a high RRI (> 0.74) corresponds with reduced diastolic flow, indicating significant renal vasoconstriction. This is a hallmark of HRS, where renal blood flow is markedly diminished despite adequate intravascular volume. In contrast, pre-renal AKI, which is characterized by hypoperfusion of the kidneys due to volume depletion, typically presents with a preserved diastolic velocity, as demonstrated in Figure 7B. These images further substantiate the diagnostic capability of Doppler imaging to identify and differentiate between pre-renal and intrinsic causes of AKI.
The visual confirmation of these waveforms underscores the clinical relevance of RRI as a reliable, real-time tool for assessing renal function and guiding therapeutic decisions. It enhances the ability to distinguish between vasoconstrictive renal injury (as in HRS) and renal dysfunction due to other causes, thereby improving the accuracy of AKI diagnosis and treatment strategies in patients with cirrhosis.
Incorporating POCUS into the phenotyping of AKI was associated with improved diagnostic precision, particularly in cases where clinical findings were ambiguous (Table 7). Among the 24 patients with unclear AKI etiology, POCUS provided valuable bedside assessment of volume status and cardiac function, aiding in the differentiation between pre-renal AKI, HRS, and ATN.
| Classification | Clinical only (n = 24) | After point-of-care ultrasound (n = 24) |
| Pre-renal | 8 | 14 |
| Hepatorenal syndrome | 6 | 10 |
| Acute tubular necrosis | 10 | 0 |
POCUS was conducted using a phased-array probe (2-5 MHz), and IVC collapsibility was quantified in the subcostal view. Using POCUS findings, 10 patients originally diagnosed with ATN were reclassified into either pre-renal AKI (6 patients) or HRS (4 patients). The IVC diameter and collapsibility index provided real-time insights into fluid status, while the presence of lung B-lines indicated pulmonary congestion, a feature commonly seen in HRS due to systemic vasodysregulation. The reclassification of all 10 initially suspected ATN cases supports the observation that ATN may be overdiagnosed in settings lacking real-time hemodynamic evaluation.
These findings underscore the importance of integrating non-invasive bedside tools like POCUS into the diagnostic workflow for AKI, especially in complex or borderline cases. By providing additional hemodynamic information, POCUS can support more accurate identification of reversible conditions like pre-renal AKI and early-stage HRS, potentially guiding more targeted treatment decisions and helping to avoid unnecessary interventions, particularly in resource-limited environments.
Staging by ICA-KDIGO criteria showed a clear gradient in unadjusted 30-day mortality across AKI stages (Table 8). Mortality rates increased from 7.1% (stage 1) to 20.0% (stage 2) and 56.0% (stage 3). Kaplan-Meier survival curves (Figure 8) confirmed significant separation across stages (log-rank P < 0.001). This unadjusted analysis affirms the prognostic weight of AKI severity in cirrhosis, with stage 3 carrying the highest short-term mortality.
| Acute kidney injury stage | n | 30-day mortality (%) | Survival (%) | Deaths (n) |
| Stage 1 | 28 | 7.1% | 92.9% | 2 |
| Stage 2 | 25 | 20.0% | 80.0% | 5 |
| Stage 3 | 25 | 56.0% | 44.0% | 14 |
These findings support the role of ICA-KDIGO staging as a practical, prognostic tool for identifying high-risk patients with decompensated cirrhosis and AKI.
Logistic regression analysis identified several factors independently associated with the development of AKI in cirrhotic patients. Notably, a MELD-Na score > 22 [odds ratio (OR) = 2.9, 95%CI: 1.5-5.6] showed the strongest association, underlining the pivotal role of hepatic function in kidney injury. Localized infection (non-septic) (OR = 2.4, 95%CI: 1.2-4.8) was also significantly associated with AKI risk, underscoring that even in the absence of sepsis, infection-related inflammation contributes meaningfully to renal decompensation in cirrhosis. Elevated bilirubin levels (> 5 mg/dL, OR = 2.1, 95%CI: 1.0-4.3) and low serum albumin (< 2.8 g/dL, OR = 2.0, 95%CI: 1.0-4.0) further contributed to AKI likelihood, reflecting impaired liver function and malnutrition. Any nephrotoxic exposure within 7 days before AKI onset (OR = 2.5, 95%CI: 1.2-5.1) was also a notable risk factor, emphasizing the importance of preventing avoidable kidney injury through careful medication management in cirrhotic patients.
These findings highlight the multifactorial nature of AKI development in cirrhosis, where both hepatic dysfunction and external factors like infection and nephrotoxic exposures play significant roles. Early recognition and management of these risk factors may aid in preventing or mitigating AKI progression.
To quantify the added value of the NGAL-RRI model over traditional creatinine-based AKI diagnosis, we calculated the net reclassification improvement (NRI) (Table 9). This analysis evaluates whether the biomarker model more accurately reclassifies patients into correct risk categories. Compared to the creatinine-only model, NGAL-RRI achieved a significant NRI of 0.41 (P < 0.001) for AKI subtype classification and 0.36 (P = 0.002) for 30-day mortality risk stratification. This suggests that the combined biomarker approach results in meaningful clinical reclassification, particularly by correctly identifying patients with intrinsic AKI who were previously misclassified as pre-renal using creatinine alone. These findings underscore the clinical benefit of adopting NGAL and RRI for diagnostic decision-making in cirrhotic AKI.
| Outcome | Net reclassification improvement (95%CI) | P value |
| Acute kidney injury subtyping | 0.41 (0.26-0.57) | < 0.001 |
| 30-day mortality risk stratification | 0.36 (0.14-0.52) | 0.002 |
Decision curve analysis (DCA) was performed to evaluate the net clinical benefit of the NGAL-RRI algorithm across a range of threshold probabilities for ICU admission (Table 10). The NGAL-RRI model demonstrated superior net benefit over both “treat all” and “treat none” strategies and outperformed the creatinine-only model at all clinically relevant thresholds (10%-40%). At a threshold probability of 25% for initiating ICU transfer, the net benefit of NGAL-RRI was 0.18 vs 0.06 for creatinine. This supports its real-world application in triage decision-making by minimizing false ICU transfers without missing high-risk cases (Figure 9).
| Threshold probability | Neutrophil gelatinase-associated lipocalin-renal resistive index net benefit | Creatinine net benefit |
| 10% | 0.12 | 0.04 |
| 25% | 0.18 | 0.06 |
| 40% | 0.20 | 0.08 |
To assess the reliability of predicted probabilities, a calibration plot was generated for the XGBoost model combining NGAL and RRI (Table 11). The model showed good calibration across risk deciles, with predicted probabilities closely matching observed outcomes. The calibration slope was 0.96 (near 1), and the Brier score was 0.12 (lower is better), indicating well-calibrated predictions. Mean predicted probability in the highest decile was 0.83, compared to 0.80 observed AKI incidence. These results confirm the model’s predictive consistency and support its potential use for risk-based decision support in cirrhotic patients.
| Risk decile | Predicted risk | Observed acute kidney injury rate |
| 1 (lowest) | 0.08 | 0.05 |
| 5 (median) | 0.42 | 0.39 |
| 10 (highest) | 0.83 | 0.80 |
| Calibration slope | 0.96 | - |
| Brier score | 0.12 |
Using our decision tree model, we evaluated the practical implications of biomarker-guided ICU triage. Among 120 patients, implementation of NGAL-first screening followed by RRI testing prevented 14 unnecessary ICU transfers (28% of total ICU recommendations under standard care) (Table 12). ICU admission was correctly allocated to 93.3% of HRS patients, while 88.6% of pre-renal cases were safely managed outside the ICU. These findings highlight the real-world benefit of biomarker-driven pathways in reducing ICU overuse and enabling more appropriate allocation of critical care resources in cirrhotic populations.
| Outcome | Neutrophil gelatinase-associated lipocalin-renal resistive index Triage | Standard care | Absolute reduction |
| Total ICU admissions | 36 | 50 | -14 |
| Correct ICU allocation (hepatorenal syndrome) | 93.3% | 86.7% | +6.6% |
| Unnecessary ICU transfers (pre-renal) | 5 | 19 | -14 |
AKI is a life-threatening complication in patients with decompensated cirrhosis, particularly in Egypt, where HCV-related cirrhosis remains endemic and healthcare resources are often limited[2]. Traditional markers like serum creatinine are unreliable in this population due to reduced hepatic production and sarcopenia, leading to delayed diagnosis and poor outcomes[3]. This study aimed to evaluate the diagnostic and prognostic utility of novel biomarkers, serum NGAL, cystatin C, and RRI, for early AKI detection, subtyping, and mortality risk stratification in cirrhotic patients. By add
In this prospective cohort of 120 patients with decompensated cirrhosis (Table 1), AKI was observed in 65% of the study population, a prevalence consistent with international estimates for hospitalized cirrhotic patients[10]. Subclassification revealed that pre-renal AKI was the most common phenotype (55.1%), while intrinsic AKI comprised 44.9%, including 25.6% with HRS and 19.2% with ATN. Furthermore, the distribution of prevalent (present at admission, 66.7%) and incident (developing post-admission, 33.3%) AKI cases was balanced across these subtypes (P = 0.55), reflecting the pragmatic diagnostic challenge faced in hepatology units. Non-AKI controls exhibited significantly lower MELD-Na scores and creatinine levels compared to those with intrinsic AKI (P < 0.001), reflecting the intertwined deterioration of hepatic and renal function in advanced disease stages. This distribution of AKI phenotypes aligns with global patterns but carries region-specific implications in Egypt, where HCV-related cirrhosis dominates and healthcare infrastructure is often under strain[11]. Unlike Western cohorts, where alcohol-induced liver disease prevails[12], Egyptian cirrhotics more frequently present with dual insults from HCV and metabolic dysfunction-associated steatotic liver disease, which may accelerate renal decompensation and alter AKI subtypes[13,14]. The predominance of pre-renal AKI in our cohort highlights the clinical importance of early hemodynamic evaluation, as timely volume expansion remains effective in reversing functional renal impairment in a substantial proportion of cirrhotic patients[15]. Moreover, the observed gradation in MELD-Na and serum creatinine levels across AKI subtypes serves not only as a marker of severity but also as a critical tool in triaging patients for interventions such as albumin infusion, vasopressor support, or RRT. Our ri
Our evaluation of three non-creatinine-based markers, serum NGAL, cystatin C, and RRI, demonstrated superior diagnostic performance compared to traditional serum creatinine (Table 2). NGAL at a threshold of > 150 ng/mL yielded the highest diagnostic accuracy for AKI (AUC of 0.91), with a sensitivity of 90% and specificity of 85%. Cystatin C also exhibited strong performance (AUC of 0.86), particularly useful in patients where muscle wasting may confound creatinine interpretation. RRI > 0.74, obtained via Doppler ultrasonography, emerged as a powerful discriminator for HRS (AUC of 0.89), surpassing both creatinine and cystatin C in subtype differentiation. By contrast, creatinine displayed a notably lower diagnostic capacity (AUC of 0.73), underscoring its limited reliability in cirrhotics. Further supporting these findings, specialized tests such as urinary granular casts and hormonal profiling (Table 3) confirmed the specificity of granular casts for ATN (100% in ATN vs 0% in HRS/pre-renal AKI) and highlighted elevated renin/aldosterone levels in HRS, reinforcing the mechanistic basis for RRI’s diagnostic utility.
The diagnostic superiority of NGAL in our cohort is consistent with prior studies demonstrating its early rise after tubular injury, preceding serum creatinine elevation[16]. Importantly, our data-derived cutoff of 150 ng/mL (determined via Youden’s index), though slightly higher than Western thresholds (typically 130-140 ng/mL), may reflect the greater systemic inflammation and baseline tubular stress seen in Egypt’s HCV-dominant cirrhotic population[17]. This region-specific threshold underscores the importance of contextual biomarker validation in high-inflammation settings.
While cystatin C demonstrates lower sensitivity compared to NGAL, it serves as a reliable estimator of GFR that is unaffected by variations in muscle mass, hepatic function, or nutritional status[18]. This characteristic makes it particularly valuable in cirrhotic patients with sarcopenia, where conventional creatinine-based GFR measurements often significantly underestimate true renal function[19]. Although cystatin C has limitations, including susceptibility to inflammation (which may elevate levels independently of GFR) and thyroid dysfunction[20], we mitigated these con
The RRI, derived from Doppler ultrasonography, provides essential hemodynamic characterization for AKI subtyping in cirrhosis. Elevated RRI values are strongly suggestive of increased intrarenal vascular resistance, a hallmark feature of HRS due to systemic vasodilation and compensatory renal vasoconstriction, as demonstrated in critical care settings with similar hemodynamic profiles[22]. Although the underlying triggers differ (e.g., sepsis in the ICU vs splanchnic vasodi
Urinary microscopy and hormonal profiling provided definitive discrimination between intrinsic AKI subtypes (Table 3). Granular casts (≥ 5/hpf), pathognomonic for ATN, were present in 100% of ATN cases (15/15) but absent in both HRS (0/20) and pre-renal AKI (0/43), confirming their 100% specificity for structural tubular injury. Complementing these findings, HRS patients exhibited significantly elevated renin-angiotensin-aldosterone system (RAAS) activation, with plasma renin (12.1 ± 4.0 ng/mL) and aldosterone (240 ± 68 pg/mL) levels surpassing those in ATN and pre-renal groups (P < 0.01). This biochemical profile mirrors the hemodynamic derangements of HRS (systemic vasodi
Granular (or ’muddy brown’) casts, formed from degenerated renal tubular epithelial cells and Tamm-Horsfall protein, serve as histological hallmarks of tubular injury. Their 100% specificity for ATN in our cohort aligns with prospective studies demonstrating equivalent diagnostic accuracy of serial urinary microscopy in AKI (sensitivity of 94%, specificity of 100% for ATN)[28]. The complete absence of these casts in HRS and pre-renal AKI robustly differentiates structural from functional renal injury, corroborating pathophysiological models where HRS manifests as hemodynamic dys
In a targeted subgroup of 28 patients with diagnostically ambiguous features, hormonal profiling provided supportive subtype discrimination, with HRS patients exhibiting marked RAAS activation (renin: 12.1 ± 4.0 ng/mL; aldosterone: 240 ± 68 pg/mL). These findings should be interpreted cautiously, given the small sample size and lack of adjustment for sodium intake or diuretic effects, but they align with the known pathophysiology of HRS. These elevations reflect the extreme neurohormonal hyperactivation characteristic of HRS, where splanchnic vasodilation triggers compensatory renin release, a response pattern mirroring acute psychosocial stress models (albeit chronically sustained in cirrhosis)[30]. The magnitude of RAAS activation in HRS far exceeded levels in ATN or pre-renal AKI (P < 0.01), paralleling ob
This dual-modality approach (urinary sediment + hormonal markers) overcomes limitations of traditional tools like FENa, which is often confounded by diuretics in cirrhosis. By integrating structural and functional biomarkers, clinicians can accurately guide therapy: Vasopressors for HRS vs supportive care for ATN[33].
Prognostic modeling identified NGAL, RRI, and AKI stage as factors independently associated with adverse outcomes (Tables 4 and 5). Elevated NGAL (> 150 ng/mL) was associated with a 3.1-fold increase in 30-day mortality risk (95%CI: 1.7-5.5; P < 0.001), while RRI > 0.74 was associated with a 2.2-fold higher risk of poor renal recovery or death (95%CI: 1.3-3.7; P = 0.003). Mortality escalated with AKI stage progression: (1) Stage 1 (7.1%; HR = 2.8, 95%CI: 0.7-10.5); (2) Stage 2 (20.0%; HR = 4.6, 95%CI: 1.5-13.8); and (3) Stage 3 (56.0%; HR = 9.3, 95%CI: 3.7-23.4; all P < 0.05 except stage 1). MELD-Na > 25 was also independently associated with increased mortality risk (HR = 3.9, 95%CI: 2.1-7.2; P = 0.01).
NGAL demonstrated both diagnostic and prognostic superiority in our cohort through its dual pathophysiological roles. As an early marker of tubular epithelial injury, NGAL is rapidly released from damaged renal tubular cells within hours of AKI onset, reflecting direct structural insult rather than delayed GFR changes. Unlike creatinine – which is confounded by reduced hepatic synthesis in cirrhosis and muscle wasting in sarcopenia – NGAL is unaffected by these limitations, enabling earlier and more reliable detection of renal dysfunction. Furthermore, NGAL functions as a sensitive marker of systemic inflammation, with levels elevated in response to pro-inflammatory cytokines such as interleukin-6 and tumor necrosis factor-alpha that drive oxidative stress and multiorgan failure in decompensated cirrhosis. This dual identity, as both a tubular injury marker and an inflammation sensor, explains NGAL’s strong association with AKI detection, subtype differentiation, and short-term mortality risk, providing a mechanistic foundation for its clinical utility in cirrhosis-associated AKI[34]. In our cohort, NGAL > 150 ng/mL was associated with a 3.1-fold increase in mortality risk, corroborating mechanistic studies that demonstrate NGAL’s role in iron-mediated free radical generation and amplification of renal and hepatic cellular injury[35]. This underlines NGAL’s utility in early prognostication, especially among sarcopenic patients, where muscle wasting renders creatinine-based assessments unreliable[36].
Similarly, the RRI offers independent prognostic value by quantifying intrarenal vascular resistance. An RRI > 0.74 suggests structural microvascular alterations and impaired diastolic perfusion, features consistent with the progression to HRS[37]. Importantly, our findings show that RRI maintains prognostic significance independent of MELD-Na scores, implying it captures discrete hemodynamic information not encompassed by liver disease severity metrics[38].
The stratified mortality rates across AKI stages (stage 1: 7.1%; stage 2: 20%; stage 3: 56%) reaffirm the utility of ICA-KDIGO criteria in decompensated cirrhosis. Notably, our observed stage 3 mortality surpassed rates reported in large Western cohorts (typically 40%-50%)[39], which may reflect delayed healthcare access and limited dialysis availability in resource-constrained settings. This discrepancy highlights the need for biomarker-integrated risk models and timely intervention algorithms tailored to low-income and middle-income countries (LMICs).
The integration of ML and POCUS was associated with improved AKI subtyping accuracy (Tables 6 and 7). Using repeated nested CV, the NGAL-RRI XGBoost model demonstrated superior internal diagnostic performance [CV-AUC: 0.92 (95%CI: 0.89-0.95); ΔCV-AUC: +0.15 vs creatinine-based models], with reclassification of 42% of initially ambiguous cases previously labeled as ATN. The model also demonstrated good calibration (calibration slope of 0.96, Brier score of 0.12), supporting its reliability in predicting subtype probabilities. Concurrently, POCUS provided additional diagnostic clarity by identifying volume-responsive pre-renal AKI (n = 6) and early HRS (n = 4) via dynamic IVC and lung ultrasound assessments, reinforcing its value in bedside hemodynamic evaluation. Notably, the model’s performance improved further when borderline FENa cases (0.1%-0.2%) were excluded, a range often confounded by diuretic use and altered tubular sodium handling in cirrhosis, supporting the observation that clearer phenotypic definitions are associated with improved ML-based classification.
ML algorithms like XGBoost offer significant advantages in modeling nonlinear relationships and uncovering complex biomarker interactions not easily captured by conventional statistical methods[40]. This dual-biomarker approach showed strong exploratory accuracy for subtyping AKI into its clinically relevant phenotypes (pre-renal, HRS, and ATN), surpassing the diagnostic resolution of traditional creatinine-based or FENa-based tools. Our internally validated ML model, based on XGBoost, achieved a cross-validated AUC of 0.92 for differentiating AKI subtypes in patients with cirrhosis. This performance, combined with good calibration metrics, suggests a promising integrated biomarker-imaging strategy. While these results are exploratory and require external validation, they represent a novel application of ML in cirrhosis-associated AKI, potentially paving the way for future automated diagnostic tools in resource-limited settings.
Our findings expand upon the work by Zheng et al[41], who applied ML models to predict renal outcomes in patients with cirrhosis and AKI, demonstrating superior prognostic performance over standard scoring systems such as MELD and SOFA. However, our study is the first, to our knowledge, to implement an NGAL-RRI-based ML classifier in a real-world, HCV-predominant setting, an environment where access to kidney biopsy and advanced diagnostic infrastructure is often limited. This has particular relevance for LMICs, where early and resource-efficient risk stratification is essential.
In parallel, POCUS proved invaluable for bedside assessment of intravascular volume status and systemic congestion, especially in diagnostically ambiguous cases. IVC collapsibility indices and lung ultrasound (B-lines) clarified fluid responsiveness and volume overload, enabling the reclassification of suspected ATN to pre-renal AKI or early HRS. These findings align with the recommendations of European Association for the Study of the Liver, who emphasized the role of bedside ultrasonography in the early assessment of renal dysfunction in decompensated cirrhosis[8].
Together, this hybrid strategy, combining ML-based biomarker analytics with dynamic bedside imaging, addresses the pathophysiological complexity and diagnostic uncertainty of cirrhosis-associated AKI. It represents a scalable, resource-sensitive approach for accurate subtyping and triage, especially in settings where both over-treatment (e.g., unnecessary ICU admission) and under-treatment (e.g., delayed vasopressor use) carry high clinical risk.
The NGAL-RRI model also demonstrated superior reclassification performance, with the NRI of 0.41 for AKI sub
Staging by ICA-KDIGO criteria (Table 8) revealed a dose-dependent increase in unadjusted 30-day mortality: (1) 7.1% (stage 1); (2) 20.0% (stage 2); and (3) 56.0% (stage 3) (P < 0.001 for trend). The corresponding adjusted HRs (Table 5) confirmed a steep mortality gradient, with stages 2 and 3 showing significant independent hazards. This gradient underscores that worsening AKI severity strongly predicts short-term mortality in cirrhosis.
These findings reinforce the prognostic significance of AKI staging in cirrhosis, extending beyond its diagnostic role to serve as a clinically actionable stratification framework. The mortality gradient observed across AKI stages in our cohort aligns with previous large-scale studies, including the European CANONIC study and subsequent ICA-AKI validations, which demonstrated similarly escalating risks of death from stage 1 through stage 3 in patients with acute decompen
Stage 1 AKI, while associated with relatively low mortality, often reflects reversible hemodynamic derangements such as volume depletion or diuretic-induced hypoperfusion. This stage may therefore represent a critical window for early intervention, favoring volume optimization and close outpatient follow-up over aggressive interventions. In contrast, stage 2 and particularly stage 3 AKI signify more severe renal and systemic involvement, often overlapping with systemic inflammation, sepsis, and multiorgan failure[43]. These stages are also more likely to require RRT, vasoconstrictor therapy in the context of HRS, and transplant evaluation when feasible.
In our Egyptian cohort, the observed 56% mortality for stage 3 AKI exceeds the 40%-50% typically reported in Western settings[44]. This disparity likely reflects delays in AKI recognition, limited access to ICU care and RRT, and broader resource constraints. Importantly, the mortality risk associated with intermediate-stage (stage 2) AKI is frequently underappreciated, particularly when clinicians rely solely on serum creatinine, a flawed metric in sarcopenic cirrhotics due to reduced muscle mass and hepatic creatinine synthesis[45]. These observations support an urgent need for biomarker-guided triage and timely escalation of care for patients with stage 2 or stage 3 AKI. Moreover, stage 2 should no longer be viewed as a benign or watchful-waiting phase, but rather a pivotal turning point warranting early thera
Multivariable logistic regression analysis identified several clinically significant factors independently associated with AKI in decompensated cirrhosis (Table 13). The strongest association was observed for MELD-Na score > 22, corresponding to 2.9-fold increased odds of AKI (95%CI: 1.5-5.6; P = 0.002). Localized, non-septic infection was also substantially associated with AKI risk (OR = 2.4; 95%CI: 1.2-4.8; P = 0.015), underscoring that even in the absence of sepsis, infection-related inflammation contributes meaningfully to renal decompensation. Indicators of advanced hepatic dysfunction were strongly linked to AKI: (1) Serum bilirubin > 5 mg/dL (OR = 2.1; 95%CI: 1.0-4.3; P = 0.049); and (2) Serum albumin < 2.8 g/dL (OR = 2.0; 95%CI: 1.0-4.0; P = 0.048). Among modifiable exposures, nephrotoxic medications significantly increased AKI likelihood (OR = 2.5; 95%CI: 1.2-5.1; P = 0.013). These findings emphasize the multifactorial nature of AKI in cirrhosis, implicating hepatic reserve, infection-associated inflammation, nutritional status, and iatro
| Variable | Odds ratio (95%CI) | P value |
| Model for End-Stage Liver Disease includes serum sodium (MELD-Na) > 22 | 2.9 (1.5-5.6) | 0.002 |
| Localized infection (non-septic) | 2.4 (1.2-4.8) | 0.015 |
| Bilirubin > 5 mg/dL | 2.1 (1.0-4.3) | 0.049 |
| Albumin < 2.8 g/dL | 2.0 (1.0-4.0) | 0.048 |
| Nephrotoxic agents | 2.5 (1.2-5.1) | 0.013 |
The strong association between MELD-Na and AKI development highlights the critical interplay of liver dysfunction and renal perfusion in cirrhosis. As a composite score, MELD-Na integrates both hepatic (bilirubin, INR) and renal (creatinine, sodium) parameters, serving as a proxy for systemic hemodynamic instability and inflammation[46]. Our identified threshold (> 22) is consistent with prior studies, including the work of Kim et al[47], which identified MELD-Na ≥ 21 as an inflection point for heightened AKI risk and increased mortality in cirrhotic cohorts.
Localized, non-septic infection emerged as a potent trigger for AKI, with an OR of 2.4 in our analysis. By excluding sepsis/septic shock, we isolated the contribution of infection-associated inflammation without the profound hemody
Markers of advanced hepatic decompensation, such as hyperbilirubinemia (> 5 mg/dL) and hypoalbuminemia (< 2.8 g/dL), were also independently associated with AKI risk. Elevated bilirubin contributes to renal tubular injury via bile cast nephropathy and oxidative stress[49], while hypoalbuminemia compromises oncotic pressure, drug binding, and antioxidant defenses, exacerbating intrarenal hypoperfusion[50]. Exposure to nephrotoxic agents within 7 days before AKI onset, including non-steroidal anti-inflammatory drugs, aminoglycosides, and proton pump inhibitors, was signi
These findings suggest that MELD-Na, infection status, and biochemical markers of hepatic decompensation should be integrated into routine inpatient AKI risk assessment. Early identification of high-risk patients enables tailored preventive strategies, such as judicious diuretic use, prompt infection control, and avoidance of nephrotoxins, that may attenuate AKI onset. In high-burden settings like Egypt, embedding these predictors into a bedside checklist could support anticipatory monitoring and individualized volume management.
In a simulated triage framework, our NGAL-RRI-based protocol demonstrated potential improvements in clinical decision-making and ICU resource allocation. It is important to note that this was a modeling exercise; prospective validation is needed to confirm its real-world impact on patient outcomes and ICU utilization. As evidenced in Tables 9, 10, 11, and 12, this combined biomarker and Doppler ultrasound approach achieved three primary outcomes of clinical importance. First, the algorithm reduced unnecessary ICU admissions by 28% (14/50 avoided) through accurate identification of patients with pre-renal AKI, 88.6% of whom were appropriately managed in non-critical care settings. Second, it optimized ICU utilization for high-risk patients by correctly triaging 93.3% of HRS cases to critical care. Third, the model showed superior risk stratification capability compared to traditional creatinine-based approaches, with the NRI of 0.41 for AKI subtyping and 0.36 for risk stratification, both statistically significant at P < 0.01.
The clinical utility of this approach was further validated through DCA presented in Figure 8, which demonstrated consistent net benefit across the full spectrum of clinically relevant decision thresholds (10%-40% probability for ICU transfer). At the particularly important 25% probability threshold – representing a balanced approach between over-triage and under-triage – the NGAL-RRI model achieved a net benefit of 0.18, representing a threefold improvement over the creatinine-only model (net benefit 0.06). This enhanced performance was maintained across all threshold probabilities, consistently outperforming both “treat all” and “treat none” strategies.
These findings hold particular significance for resource-constrained clinical environments, where the combination of rapid NGAL testing and readily available Doppler ultrasonography provides an accurate yet practical diagnostic approach that circumvents the limitations of more complex or delayed testing modalities[51]. The protocol’s ability to simultaneously improve diagnostic accuracy while optimizing use of limited ICU resources represents an important advancement in the management of cirrhosis-associated AKI.
These findings demonstrate that biomarker-guided decision-making can meaningfully enhance both clinical outcomes and ICU resource utilization in decompensated cirrhosis. NGAL, an early marker of tubular stress and systemic inflammation, enables rapid detection of intrinsic AKI before creatinine levels rise[52]. When combined with RRI, a hemo
The use of DCA adds significant translational strength. Unlike traditional accuracy metrics, DCA evaluates net clinical benefit across varying ICU referral thresholds, balancing the harms of over-triage and under-triage. At a 25% probability threshold, a clinically relevant inflection point, our model delivered a net benefit of 0.18, threefold greater than creatinine-based models (0.06). This suggests that biomarker integration could materially improve ICU triage precision, particularly in resource-limited settings like Egypt, where critical care availability is constrained[54]. Furthermore, calibration analysis confirmed the NGAL-RRI model’s predictive reliability, with close alignment between predicted and observed outcomes (calibration slope = 0.96, Brier score = 0.12, Table 11), underscoring its generalizability to real-world hepatology units. Collectively, these results advocate for the implementation of this combined biomarker-Doppler framework as a practical and scalable triage tool in cirrhosis-associated AKI.
The NGAL-RRI protocol offers a low-cost, high-impact triage strategy for cirrhotic AKI, improving diagnostic precision and ICU resource allocation. Its bedside applicability supports more rational ICU referrals, timely escalation for high-risk cases, and conservative management of low-risk patients.
The findings of this study support the integration of NGAL and RRI as early and actionable biomarkers for the detection, subtyping, and management of AKI in patients with decompensated cirrhosis. This dual-marker approach enables real-time therapeutic decision-making across the AKI continuum: Guiding timely volume resuscitation in pre-renal states, facilitating the safe avoidance of nephrotoxins, and prompting early initiation of vasoconstrictor therapy in patients with evolving HRS. NGAL-guided risk stratification additionally enhances critical care triage, reducing unnecessary ICU admissions and conserving intensive care resources, a critical advantage in resource-constrained settings. Concurrently, RRI offers a non-invasive, bedside method to assess renal vascular tone and tailor hemodynamic management, including fluid responsiveness and vasopressor titration. While slightly less sensitive, cystatin C con
Collectively, the NGAL-RRI algorithm offers a scalable, resource-efficient solution that aligns with the clinical and logistical constraints of LMICs, including Egypt. Its implementation may serve as a pragmatic pathway to improve AKI outcomes through earlier diagnosis, individualized intervention, and more efficient resource allocation.
This study represents the first prospective diagnostic investigation in an HCV-endemic, resource-limited setting to systematically evaluate the combined utility of NGAL, cystatin C, and Doppler-derived RRI for early AKI detection, subtype differentiation (pre-renal, HRS, ATN), and short-term prognostication in cirrhosis. It is also the first to integrate biochemical, hemodynamic, and imaging parameters into a unified diagnostic framework specifically optimized for cirrhotic care, marking a decisive shift beyond sole reliance on serum creatinine. The incorporation of a ML classifier (XGBoost) to enhance AKI subtyping represents a methodological advance, yielding high diagnostic accuracy and enabling personalized risk assessment. Importantly, the inclusion of real-world clinical decision tools, such as NGAL-guided ICU triage and DCA, translates these findings into actionable bedside strategies. By deliberately excluding borderline FENa cases, the study also addresses a persistent challenge in AKI phenotyping, thereby enhancing diagnostic precision. Taken together, these contributions establish a scalable, biomarker-driven model for AKI diagnosis and management in cirrhosis, with particular relevance to LMICs where diagnostic resources are often constrained.
Several limitations should be acknowledged. First, the study was conducted at a single tertiary care center, which may affect the external generalizability of findings across diverse populations and healthcare settings. Although the sample size was adequate for diagnostic performance analysis, it was not powered to detect granular differences in subgroup-specific mortality or long-term renal outcomes. Second, while the XGBoost model demonstrated strong internal validity via repeated nested CV, it has not been externally validated in an independent cohort, which limits generalizability until further validation is performed. Third, while we excluded sepsis/septic shock using sepsis-3 criteria to minimize con
While this study establishes the diagnostic utility of NGAL-RRI in cirrhotic AKI, several important questions remain for future investigation. First, multicenter validation across diverse resource-limited settings is needed to confirm generalizability, particularly in HCV-endemic regions with emerging metabolic liver diseases. Second, the development of point-of-care NGAL assays could address resource limitations while maintaining diagnostic accuracy. Third, in
We recognize the practical challenge of incorporating advanced diagnostics into time-limited clinical encounters. Although this study employed comprehensive biomarker and imaging assessment to define AKI phenotypes with high precision, the clinical relevance of these findings lies in their translation into a focused and efficient diagnostic strategy. Our results support a stepped-care framework in which investigations are prioritized according to their immediate impact on management, rather than applied indiscriminately.
Within this framework, serum NGAL functions as an early triage tool at the initial suspicion of AKI. A single NGAL measurement allows rapid risk stratification within hours, substantially earlier than creatinine-based trends. Low NGAL levels favor a functional or pre-renal process and support continued ward-based management with targeted hemo
This diagnostic clarity is of paramount importance in specific high-stakes clinical scenarios. A prime example is the management of cirrhotic patients undergoing surveillance or treatment for HCC. In this population, AKI may reflect hemodynamic dysfunction (e.g., HRS), therapy-related nephrotoxicity (e.g., from tyrosine kinase inhibitors or immunotherapy), or intrinsic renal injury (e.g., ATN). Misdiagnosis of functional HRS as intrinsic ATN could lead to unnecessary and harmful cessation of oncological therapy. Conversely, incorrectly attributing AKI to cirrhosis when it is drug-induced ATN risks continued nephrotoxic exposure and irreversible renal damage. The NGAL-guided and RRI-guided algorithm provides a rapid, non-invasive means to distinguish between these critical etiologies, supporting informed, multidisciplinary decision-making between hepatology and oncology teams to optimize both hepatic and oncologic outcomes.
Only a minority of patients require escalation to additional confirmatory investigations when biomarker and imaging findings are inconclusive. By structuring diagnostics sequentially in this manner, the proposed framework reallocates clinical time from managing the consequences of delayed diagnosis toward making earlier, more definitive treatment decisions, aligning diagnostic intensity with clinical need.
Cirrhotic patients undergoing surveillance or treatment for HCC represent a distinct clinical subgroup in whom AKI frequently has multifactorial causes[46,48]. In these patients, renal dysfunction may reflect hemodynamic changes related to cirrhosis progression, treatment-related nephrotoxicity, or intrinsic kidney injury[8,29]. Misclassification of functional kidney injury as structural disease may lead to unnecessary interruption of oncological therapy, whereas failure to recognize true intrinsic injury risks continued nephrotoxic exposure[33,37]. Early etiologic differentiation using an NGAL-guided and RRI-guided approach provides a rapid, non-invasive means to clarify the predominant mechanism of AKI, supporting informed, multidisciplinary decision-making between hepatology and oncology teams and helping to preserve both renal function and continuity of cancer care[17,20].
Our findings hold particular relevance for LMICs, such as Egypt, where access to advanced diagnostics, ICU beds, and RRT is often constrained. The combined use of NGAL (via commercially available ELISA kits) and RRI (via portable Doppler ultrasound) offers a relatively low-cost, bedside-compatible diagnostic strategy that avoids reliance on delayed or confounded markers like creatinine and FENa. The stepped-care framework outlined above makes this strategy feasible, as it concentrates resources on high-yield tests for high-risk patients. In settings where diuretic use is widespread and sarcopenia is common, this approach may improve early AKI recognition and reduce unnecessary ICU referrals. Future implementation studies should assess the cost-effectiveness, training requirements, and workflow integration of this biomarker-guided algorithm in routine hepatology practice in resource-limited environments.
Although a formal cost-effectiveness analysis, including direct monetary cost calculations, was beyond the scope of the present diagnostic accuracy study and no cost data were prospectively collected, the observed reduction in ICU utilization provides a strong economic rationale for early biomarker-guided and POCUS-guided triage. By preventing inappropriate vasoconstrictor use, excessive fluid administration, and delayed recognition of intrinsic kidney injury, this approach may also reduce downstream complications and long-term chronic kidney disease burden. In resource-constrained healthcare systems, targeted investment in NGAL testing and focused POCUS training may therefore translate into clinically meaningful and economically sustainable improvements in care delivery.
NGAL and RRI significantly outperform serum creatinine in the early diagnosis and phenotyping of AKI in cirrhotic patients. NGAL is reliably associated with 30-day mortality, while RRI identifies patients at risk of poor renal recovery, including those with HRS. Cystatin C adds diagnostic clarity in borderline or ambiguous presentations. An integrated NGAL-RRI approach, supported by cystatin C when needed, enables precise, timely AKI subtyping and prognostication. This strategy represents a scalable, evidence-based framework for early AKI detection and triage in cirrhotic populations, particularly in resource-limited settings. By facilitating smarter ICU resource allocation and individualized therapy, adoption of this biomarker-guided pathway may improve clinical outcomes in cirrhosis-associated AKI and reduce reliance on unreliable creatinine-based criteria.
| 1. | Angeli P, Garcia-Tsao G, Nadim MK, Parikh CR. News in pathophysiology, definition and classification of hepatorenal syndrome: A step beyond the International Club of Ascites (ICA) consensus document. J Hepatol. 2019;71:811-822. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 375] [Cited by in RCA: 310] [Article Influence: 44.3] [Reference Citation Analysis (1)] |
| 2. | Jean K, Tawheed A, Luong Nguyen LB, Heikal T, Eldaly U, Elhadidy N, Elghaieb A, Aboudonia A, Tondeur L, Dublineau A, Fontanet A, El-Kassas M. Changes in Presentation, Treatment, and Survival of Patients with Hepatocellular Carcinoma in Damietta, Egypt, 2007-2019: A Retrospective Monocentric Cohort Study. J Hepatocell Carcinoma. 2023;10:99-111. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 4] [Cited by in RCA: 8] [Article Influence: 2.7] [Reference Citation Analysis (0)] |
| 3. | Br VK, Sarin SK. Acute-on-chronic liver failure: Terminology, mechanisms and management. Clin Mol Hepatol. 2023;29:670-689. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 103] [Cited by in RCA: 85] [Article Influence: 28.3] [Reference Citation Analysis (2)] |
| 4. | Freiz M, Mohamed A, Mousa Y. Biomarkers for Acute Kidney Injury. MJMR. 2022;46-49. [DOI] [Full Text] |
| 5. | Larstorp ACK, Salvador CL, Svensvik BA, Klingenberg O, Distante S. Neutrophil Gelatinase-Associated Lipocalin (NGAL) and cystatin C are early biomarkers of acute kidney injury associated with cardiac surgery. Scand J Clin Lab Invest. 2022;82:410-418. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 10] [Reference Citation Analysis (0)] |
| 6. | Huang DQ, Terrault NA, Tacke F, Gluud LL, Arrese M, Bugianesi E, Loomba R. Global epidemiology of cirrhosis - aetiology, trends and predictions. Nat Rev Gastroenterol Hepatol. 2023;20:388-398. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 666] [Cited by in RCA: 600] [Article Influence: 200.0] [Reference Citation Analysis (5)] |
| 7. | Khairallah MK, Kellani A, Ibrahim HM, Ahmed MS, Mostafa MH. Role of Renal Doppler Ultrasound in Early Detection of Acute Kidney Injury in Critically Ill Patients. Egypt J Hosp Med. 2022;88:2661-2667. [DOI] [Full Text] |
| 8. | European Association for the Study of the Liver. EASL Clinical Practice Guidelines for the management of patients with decompensated cirrhosis. J Hepatol. 2018;69:406-460. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 2259] [Cited by in RCA: 2114] [Article Influence: 264.3] [Reference Citation Analysis (8)] |
| 9. | K P, A A, Reddy SY. Renal Resistive Index as a Prognostic Indicator in Patients With Hepatic Cirrhosis: An Analytical Cross-Sectional Study. Cureus. 2025;17:e82700. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 1] [Reference Citation Analysis (0)] |
| 10. | Desai AP, Knapp SM, Orman ES, Ghabril MS, Nephew LD, Anderson M, Ginès P, Chalasani NP, Patidar KR. Changing epidemiology and outcomes of acute kidney injury in hospitalized patients with cirrhosis - a US population-based study. J Hepatol. 2020;73:1092-1099. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 79] [Cited by in RCA: 82] [Article Influence: 13.7] [Reference Citation Analysis (0)] |
| 11. | Vaara ST, Bhatraju PK, Stanski NL, McMahon BA, Liu K, Joannidis M, Bagshaw SM. Subphenotypes in acute kidney injury: a narrative review. Crit Care. 2022;26:251. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 46] [Cited by in RCA: 77] [Article Influence: 19.3] [Reference Citation Analysis (0)] |
| 12. | Charatcharoenwitthaya P, Liangpunsakul S, Piratvisuth T. Alcohol-Associated Liver Disease: East Versus West. Clin Liver Dis (Hoboken). 2020;16:231-235. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 10] [Cited by in RCA: 10] [Article Influence: 1.7] [Reference Citation Analysis (0)] |
| 13. | El Behery M, Elghwab AI, Tabll A, El-Sayed EH, Abdelrazek MA. Review Article: Egyptian Efforts to Control Hepatitis C Virus in Egypt. AJBAS. 2025;6:253-262. [DOI] [Full Text] |
| 14. | Elhoseeny MM, Rageh F, Rezk SM, Othman AAA. Frequency and risk factors of metabolic associated fatty liver disease among medical students in Egypt. Sci Rep. 2025;15:13470. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 2] [Cited by in RCA: 6] [Article Influence: 6.0] [Reference Citation Analysis (0)] |
| 15. | Sanchez LO, Francoz C. Global strategy for the diagnosis and management of acute kidney injury in patients with liver cirrhosis. United European Gastroenterol J. 2021;9:220-228. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 7] [Cited by in RCA: 6] [Article Influence: 1.2] [Reference Citation Analysis (0)] |
| 16. | Slocum JL, Heung M, Pennathur S. Marking renal injury: can we move beyond serum creatinine? Transl Res. 2012;159:277-289. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 94] [Cited by in RCA: 100] [Article Influence: 7.1] [Reference Citation Analysis (0)] |
| 17. | Belcher JM, Garcia-Tsao G, Sanyal AJ, Thiessen-Philbrook H, Peixoto AJ, Perazella MA, Ansari N, Lim J, Coca SG, Parikh CR; TRIBE-AKI Consortium. Urinary biomarkers and progression of AKI in patients with cirrhosis. Clin J Am Soc Nephrol. 2014;9:1857-1867. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 90] [Cited by in RCA: 84] [Article Influence: 7.0] [Reference Citation Analysis (0)] |
| 18. | Laterza OF, Price CP, Scott MG. Cystatin C: an improved estimator of glomerular filtration rate? Clin Chem. 2002;48:699-707. [PubMed] [DOI] [Full Text] |
| 19. | Mindikoglu AL, Dowling TC, Weir MR, Seliger SL, Christenson RH, Magder LS. Performance of chronic kidney disease epidemiology collaboration creatinine-cystatin C equation for estimating kidney function in cirrhosis. Hepatology. 2014;59:1532-1542. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 106] [Cited by in RCA: 104] [Article Influence: 8.7] [Reference Citation Analysis (1)] |
| 20. | Belcher JM, Sanyal AJ, Peixoto AJ, Perazella MA, Lim J, Thiessen-Philbrook H, Ansari N, Coca SG, Garcia-Tsao G, Parikh CR; TRIBE-AKI Consortium. Kidney biomarkers and differential diagnosis of patients with cirrhosis and acute kidney injury. Hepatology. 2014;60:622-632. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 283] [Cited by in RCA: 259] [Article Influence: 21.6] [Reference Citation Analysis (1)] |
| 21. | Stevens LA, Coresh J, Schmid CH, Feldman HI, Froissart M, Kusek J, Rossert J, Van Lente F, Bruce RD 3rd, Zhang YL, Greene T, Levey AS. Estimating GFR using serum cystatin C alone and in combination with serum creatinine: a pooled analysis of 3,418 individuals with CKD. Am J Kidney Dis. 2008;51:395-406. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 841] [Cited by in RCA: 806] [Article Influence: 44.8] [Reference Citation Analysis (7)] |
| 22. | Cuttone G, Geraci G, La Via L, Sorbello M, Pappalardo F, Carollo C. Exploring the Utility of Renal Resistive Index in Critical Care: Insights into ARDS and Cardiac Failure. Biomedicines. 2025;13:519. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 5] [Reference Citation Analysis (0)] |
| 23. | Nour I. Relation between renal resistive index and renal function in septic patients developing acute kidney injury 2. Egypt J Intensive Care Emerg Med. 2022;2:21-28. [DOI] [Full Text] |
| 24. | Surya H, Kumar R, Priyadarshi RN, Surya Prakash S, Kumar S. Renal resistive index measurements by ultrasound in patients with liver cirrhosis: Magnitude and associations with renal dysfunction. World J Radiol. 2024;16:221-231. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in CrossRef: 2] [Cited by in RCA: 3] [Article Influence: 1.5] [Reference Citation Analysis (2)] |
| 25. | Mogawer MS, Nassef SAR, Elhamid SMA, Elkholy S, El Aziz NEA, Al-Jarhi UM, Abdellatif AA. Role of renal Duplex ultrasonography in evaluation of hepatorenal syndrome. Egypt Liver J. 2021;11:34. [RCA] [DOI] [Full Text] [Cited by in RCA: 5] [Reference Citation Analysis (0)] |
| 26. | Toledo C, Thomas G, Schold JD, Arrigain S, Gornik HL, Nally JV, Navaneethan SD. Renal resistive index and mortality in chronic kidney disease. Hypertension. 2015;66:382-388. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 39] [Cited by in RCA: 56] [Article Influence: 5.1] [Reference Citation Analysis (0)] |
| 27. | Diab AM, Abdelmotaleb GS, Rashed MF, Abd Elateef W, Elshaarawy FS, Affi WE. Value of Serum NGAL Combined with Serum Cystatin C for Predicting Acute Kidney Injury in Preterm Neonates with Respiratory Distress Syndrome. Egypt J Hosp Med. 2022;89:6895-6901. [DOI] [Full Text] |
| 28. | Varghese V, Rivera MS, Alalwan AA, Alghamdi AM, Gonzalez ME, Velez JCQ. Diagnostic Utility of Serial Microscopic Examination of the Urinary Sediment in Acute Kidney Injury. Kidney360. 2021;2:182-191. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 6] [Cited by in RCA: 13] [Article Influence: 2.6] [Reference Citation Analysis (0)] |
| 29. | Simonetto DA, Gines P, Kamath PS. Hepatorenal syndrome: pathophysiology, diagnosis, and management. BMJ. 2020;370:m2687. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 175] [Cited by in RCA: 153] [Article Influence: 25.5] [Reference Citation Analysis (0)] |
| 30. | Tibi S, Zeynalvand G, Mohsin H. Role of the Renin Angiotensin Aldosterone System in the Pathogenesis of Sepsis-Induced Acute Kidney Injury: A Systematic Review. J Clin Med. 2023;12:4566. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 23] [Cited by in RCA: 17] [Article Influence: 5.7] [Reference Citation Analysis (0)] |
| 31. | Gideon A, Sauter C, Fieres J, Berger T, Renner B, Wirtz PH. Kinetics and Interrelations of the Renin Aldosterone Response to Acute Psychosocial Stress: A Neglected Stress System. J Clin Endocrinol Metab. 2020;105:e762-e773. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 15] [Cited by in RCA: 33] [Article Influence: 5.5] [Reference Citation Analysis (0)] |
| 32. | Shoemaker R, Poglitsch M, Huang H, Vignes K, Srinivasan A, Cockerham C, Schadler A, Bauer JA, O'Brien JM. Activation of the Renin-Angiotensin-Aldosterone System Is Attenuated in Hypertensive Compared with Normotensive Pregnancy. Int J Mol Sci. 2023;24:12728. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 18] [Reference Citation Analysis (0)] |
| 33. | Nadim MK, Kellum JA, Forni L, Francoz C, Asrani SK, Ostermann M, Allegretti AS, Neyra JA, Olson JC, Piano S, VanWagner LB, Verna EC, Akcan-Arikan A, Angeli P, Belcher JM, Biggins SW, Deep A, Garcia-Tsao G, Genyk YS, Gines P, Kamath PS, Kane-Gill SL, Kaushik M, Lumlertgul N, Macedo E, Maiwall R, Marciano S, Pichler RH, Ronco C, Tandon P, Velez JQ, Mehta RL, Durand F. Acute kidney injury in patients with cirrhosis: Acute Disease Quality Initiative (ADQI) and International Club of Ascites (ICA) joint multidisciplinary consensus meeting. J Hepatol. 2024;81:163-183. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 93] [Cited by in RCA: 157] [Article Influence: 78.5] [Reference Citation Analysis (1)] |
| 34. | Allegretti AS, Parada XV, Endres P, Zhao S, Krinsky S, St Hillien SA, Kalim S, Nigwekar SU, Flood JG, Nixon A, Simonetto DA, Juncos LA, Karakala N, Wadei HM, Regner KR, Belcher JM, Nadim MK, Garcia-Tsao G, Velez JCQ, Parikh SM, Chung RT; HRS-HARMONY study investigators. Urinary NGAL as a Diagnostic and Prognostic Marker for Acute Kidney Injury in Cirrhosis: A Prospective Study. Clin Transl Gastroenterol. 2021;12:e00359. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 104] [Cited by in RCA: 101] [Article Influence: 20.2] [Reference Citation Analysis (0)] |
| 35. | Medić B, Rovcanin B, Vujovic KS, Obradovic D, Duric D, Prostran M. Evaluation of Novel Biomarkers of Acute Kidney Injury: The Possibilities and Limitations. Curr Med Chem. 2016;23:1981-1997. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 25] [Cited by in RCA: 37] [Article Influence: 3.7] [Reference Citation Analysis (0)] |
| 36. | Post A, Groothof D, Eisenga MF, Bakker SJL. Sodium-Glucose Cotransporter 2 Inhibitors and Kidney Outcomes: True Renoprotection, Loss of Muscle Mass or Both? J Clin Med. 2020;9:1603. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in Crossref: 7] [Cited by in RCA: 19] [Article Influence: 3.2] [Reference Citation Analysis (0)] |
| 37. | Pose E, Piano S, Juanola A, Ginès P. Hepatorenal Syndrome in Cirrhosis. Gastroenterology. 2024;166:588-604.e1. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 43] [Reference Citation Analysis (1)] |
| 38. | Narayanan V, Devadas K, Sreesh S, Varghese J, Solanki R, Mohapatra SD, Pal R, Madhu D, Chakravorty A. Novel predictors of response to therapy with terlipressin and albumin in hepatorenal syndrome-acute kidney injury. Ann Gastroenterol. 2024;37:81-88. [RCA] [PubMed] [DOI] [Full Text] [Full Text (PDF)] [Cited by in RCA: 2] [Reference Citation Analysis (0)] |
| 39. | Patidar KR, Ma AT, Juanola A, Barone A, Incicco S, Kulkarni AV, Hernández JLP, Wentworth B, Asrani SK, Alessandria C, Abdelkader NA, Wong YJ, Xie Q, Pyrsopoulos NT, Kim SE, Fouad Y, Torre A, Cerda E, Ferrer JD, Maiwall R, Simonetto DA, Papp M, Orman ES, Perricone G, Solé C, Lange CM, Farias AQ, Pereira G, Gadano A, Caraceni P, Thevenot T, Verma N, Kim JH, Vorobioff JD, Cordova-Gallardo J, Ivashkin V, Roblero JP, Maan R, Toledo C, Gioia S, Fassio E, Marino M, Nabilou P, Vargas V, Merli M, Goncalves LL, Rabinowich L, Krag A, Balcar L, Montes P, Mattos AZ, Bruns T, Mohammed A, Laleman W, Carrera E, Cabrera MC, Girala M, Samant H, Raevens S, Madaleno J, Kim RW, Arab JP, Presa J, Ferreira CN, Galante A, Allegretti AS, Takkenberg B, Marciano S, Sarin SK, Durand F, Ginès P, Angeli P, Solà E, Piano S; International Club of Ascites GLOBAL AKI team. Global epidemiology of acute kidney injury in hospitalised patients with decompensated cirrhosis: the International Club of Ascites GLOBAL AKI prospective, multicentre, cohort study. Lancet Gastroenterol Hepatol. 2025;10:418-430. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 5] [Cited by in RCA: 15] [Article Influence: 15.0] [Reference Citation Analysis (0)] |
| 40. | Chen T, Guestrin C. XGBoost: A Scalable Tree Boosting System. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining; 2016 August 13; NY, United States. United States: Association for Computing Machinery, 2016: 785-794. [RCA] [DOI] [Full Text] [Cited by in Crossref: 12755] [Cited by in RCA: 10473] [Article Influence: 1047.3] [Reference Citation Analysis (2)] |
| 41. | Zheng L, Yang J, Zhao L, Li C, Fang K, Li S, Wu J, Zheng M. Development and validation of the PHM-CPA model to predict in-hospital mortality for cirrhotic patients with acute kidney injury. Dig Liver Dis. 2025;57:485-493. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 3] [Cited by in RCA: 3] [Article Influence: 3.0] [Reference Citation Analysis (0)] |
| 42. | Angeli P, Gines P, Wong F, Bernardi M, Boyer TD, Gerbes A, Moreau R, Jalan R, Sarin SK, Piano S, Moore K, Lee SS, Durand F, Salerno F, Caraceni P, Kim WR, Arroyo V, Garcia-Tsao G; International Club of Ascites. Diagnosis and management of acute kidney injury in patients with cirrhosis: revised consensus recommendations of the International Club of Ascites. Gut. 2015;64:531-537. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 484] [Cited by in RCA: 421] [Article Influence: 38.3] [Reference Citation Analysis (0)] |
| 43. | Piano S, Rosi S, Maresio G, Fasolato S, Cavallin M, Romano A, Morando F, Gola E, Frigo AC, Gatta A, Angeli P. Evaluation of the Acute Kidney Injury Network criteria in hospitalized patients with cirrhosis and ascites. J Hepatol. 2013;59:482-489. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 240] [Cited by in RCA: 213] [Article Influence: 16.4] [Reference Citation Analysis (2)] |
| 44. | Wong F, Pappas SC, Curry MP, Reddy KR, Rubin RA, Porayko MK, Gonzalez SA, Mumtaz K, Lim N, Simonetto DA, Sharma P, Sanyal AJ, Mayo MJ, Frederick RT, Escalante S, Jamil K; CONFIRM Study Investigators. Terlipressin plus Albumin for the Treatment of Type 1 Hepatorenal Syndrome. N Engl J Med. 2021;384:818-828. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 412] [Cited by in RCA: 354] [Article Influence: 70.8] [Reference Citation Analysis (2)] |
| 45. | Carrero JJ, Johansen KL, Lindholm B, Stenvinkel P, Cuppari L, Avesani CM. Screening for muscle wasting and dysfunction in patients with chronic kidney disease. Kidney Int. 2016;90:53-66. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 153] [Cited by in RCA: 216] [Article Influence: 21.6] [Reference Citation Analysis (0)] |
| 46. | Biggins SW, Angeli P, Garcia-Tsao G, Ginès P, Ling SC, Nadim MK, Wong F, Kim WR. Diagnosis, Evaluation, and Management of Ascites, Spontaneous Bacterial Peritonitis and Hepatorenal Syndrome: 2021 Practice Guidance by the American Association for the Study of Liver Diseases. Hepatology. 2021;74:1014-1048. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 690] [Cited by in RCA: 606] [Article Influence: 121.2] [Reference Citation Analysis (3)] |
| 47. | Kim WR, Biggins SW, Kremers WK, Wiesner RH, Kamath PS, Benson JT, Edwards E, Therneau TM. Hyponatremia and mortality among patients on the liver-transplant waiting list. N Engl J Med. 2008;359:1018-1026. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 1236] [Cited by in RCA: 1124] [Article Influence: 62.4] [Reference Citation Analysis (5)] |
| 48. | Moreau R, Jalan R, Gines P, Pavesi M, Angeli P, Cordoba J, Durand F, Gustot T, Saliba F, Domenicali M, Gerbes A, Wendon J, Alessandria C, Laleman W, Zeuzem S, Trebicka J, Bernardi M, Arroyo V; CANONIC Study Investigators of the EASL–CLIF Consortium. Acute-on-chronic liver failure is a distinct syndrome that develops in patients with acute decompensation of cirrhosis. Gastroenterology. 2013;144:1426-1437, 1437.e1. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 2562] [Cited by in RCA: 2332] [Article Influence: 179.4] [Reference Citation Analysis (6)] |
| 49. | Fickert P, Krones E, Pollheimer MJ, Thueringer A, Moustafa T, Silbert D, Halilbasic E, Yang M, Jaeschke H, Stokman G, Wells RG, Eller K, Rosenkranz AR, Eggertsen G, Wagner CA, Langner C, Denk H, Trauner M. Bile acids trigger cholemic nephropathy in common bile-duct-ligated mice. Hepatology. 2013;58:2056-2069. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 147] [Cited by in RCA: 139] [Article Influence: 10.7] [Reference Citation Analysis (1)] |
| 50. | Garcia-Martinez R, Caraceni P, Bernardi M, Gines P, Arroyo V, Jalan R. Albumin: pathophysiologic basis of its role in the treatment of cirrhosis and its complications. Hepatology. 2013;58:1836-1846. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 389] [Cited by in RCA: 346] [Article Influence: 26.6] [Reference Citation Analysis (3)] |
| 51. | Helal NE, Elkadeem M, Elbastawesy SM, Mashaal SA. Drug induced liver injury: causative agents and predictors for the outcome - a retrospective study at Tanta University Hospital, Egypt. Eur Rev Med Pharmacol Sci. 2023;27:560-569. [RCA] [PubMed] [DOI] [Full Text] [Cited by in RCA: 2] [Reference Citation Analysis (0)] |
| 52. | Haase M, Bellomo R, Devarajan P, Schlattmann P, Haase-Fielitz A; NGAL Meta-analysis Investigator Group. Accuracy of neutrophil gelatinase-associated lipocalin (NGAL) in diagnosis and prognosis in acute kidney injury: a systematic review and meta-analysis. Am J Kidney Dis. 2009;54:1012-1024. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 1048] [Cited by in RCA: 922] [Article Influence: 54.2] [Reference Citation Analysis (1)] |
| 53. | Carrier P, Debette-Gratien M, Essig M, Loustaud-Ratti V. Beyond serum creatinine: which tools to evaluate renal function in cirrhotic patients? Hepatol Res. 2018;48:771-779. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 8] [Cited by in RCA: 10] [Article Influence: 1.3] [Reference Citation Analysis (0)] |
| 54. | Wong F. Recent advances in our understanding of hepatorenal syndrome. Nat Rev Gastroenterol Hepatol. 2012;9:382-391. [RCA] [PubMed] [DOI] [Full Text] [Cited by in Crossref: 71] [Cited by in RCA: 75] [Article Influence: 5.4] [Reference Citation Analysis (0)] |