Diagnostic and prognostic utility of serum bile acid in liver transplant recipients: A systematic review and quantitative synthesis

Abstract

BACKGROUND

Liver transplantation (LT) remains the definitive treatment for patients with acute and chronic end-stage liver disease, significantly improving survival and quality of life. Conventional liver function tests post LT, while routinely used, often lack sensitivity and may not detect graft injury promptly. Serum bile acids (SBAs), known for their role in hepatic function and enterohepatic circulation, have emerged as promising biomarkers due to their sensitivity to hepatocellular injury and cholestasis. This systematic review evaluates the prognostic value of SBA levels following LT.

AIM

To assess correlations between SBA concentrations and clinical outcomes, including graft function and rejection episodes.

METHODS

Relevant studies analyzing SBA levels after LT were systematically reviewed. Data synthesis focused on the association between bile acid levels and transplant outcomes.

RESULTS

Elevated SBA levels were significantly associated with early graft dysfunction and acute rejection. Pooled analysis indicated that patients with SBA levels exceeding study-specific thresholds had a 2.5-fold increased risk of acute rejection (95%CI: 1.8-3.4, P < 0.001). Diagnostic accuracy analysis showed that SBA levels had a sensitivity of 82% and specificity of 76% for predicting graft dysfunction within the first month after transplantation. Subgroup analyses highlighted conjugated bile acids as particularly predictive, with an odds ratio of 3.1 (95%CI: 2.0-4.8, P < 0.0001) for adverse outcomes.

CONCLUSION

SBA levels demonstrate strong potential as a non-invasive early biomarker for post-liver transplant prognosis, facilitating timely clinical interventions. However, heterogeneity in assay techniques and cutoff values across studies underscores the need for standardized measurement protocols. Incorporating SBA monitoring into routine post-transplant surveillance may enhance prognostic precision and improve patient management strategies.

Key Words: Serum bile acid; Liver transplantation; Surveillance strategies; Graft dysfunction; Graft rejection; Biliary complications

Core Tip: Serum bile acid (SBA) levels serve as a sensitive, non-invasive biomarker for early detection and prognosis of graft dysfunction, acute rejection, patient survival, and biliary complications following liver transplantation. Elevated SBA levels correlate strongly with adverse outcomes, often preceding changes in conventional liver function tests. Despite promising diagnostic accuracy, heterogeneity in assay methods and cutoff values limits routine clinical use. Standardized protocols and large multicenter prospective studies are needed to validate SBA monitoring as a complementary tool to established biochemical tests, enhancing post-transplant surveillance and timely intervention strategies.

INTRODUCTION

Liver transplantation (LT) is a common therapeutic modality for patients with acute and chronic end-stage liver disease. This procedure has been reported to restore normal health and lifestyle and extend the patient’s lifespan by 15 years[1]. Moreover, data from the Scientific Registry of Transplant Recipients have shown that overall patient survival is excellent, reaching 90% and 77% at 1 and 5 years after deceased donor LT, respectively[2]. Since the first attempt at LT in 1963, there have been continuous advances and major improvements in surgical technique, immunosuppression, and perioperative care. However, early graft dysfunction (EAD), acute rejection, and biliary complications continue to substantially contribute to morbidity and mortality after transplantation[3,4]. Therefore, identifying patients at risk of adverse outcomes is crucial in guiding clinical decision-making and optimizing post-transplant surveillance strategies.

Previous studies have investigated several biomarkers for predicting graft dysfunction, graft rejection, patient survival, and complications after LT. For instance, Aziz et al[5] found that aspartate aminotransferase (AST), total bilirubin, direct bilirubin, γ-glutamyl transferase (GGT), and the international normalized ratio (INR) measured between days 1 and 4 after living donor LT were independently associated with graft rejection at 3 months. Additionally, alanine aminotransferase (ALT), AST, lactate dehydrogenase, INR, C-reactive protein, uric acid, and the neutrophil-to-lymphocyte ratio measured during the same period were identified as valuable predictors of 3-month post-transplant survival. A meta-analysis of 21 studies also found that biomarkers measured within 15 days after LT were significantly associated with graft-related outcomes[6]. Nevertheless, there is no global consensus on the biomarkers that can be used for the early detection of graft rejection, graft dysfunction, or patient survival. Therefore, there is continued interest in identifying biomarkers that may provide earlier prognostic information.

Serum bile acid (SBAs) have been extensively used as sensitive markers for liver function, and preliminary studies suggest that they may help monitor clinical outcomes after LT. Indeed, an experimental study performed in 20 pigs found that normalization of total SBA levels within a few hours after LT was indicative of a well perfused and well-preserved graft, suggesting that total SBAs are a more sensitive and specific indicator of early graft function than conventional biomarkers, such as bilirubin, AST, ALT, and lactic acid[7]. Another experimental study involving immunosuppressed pigs found that SBAs were superior to standard liver function tests (LFT) in detecting early graft rejection[8]. Although these findings indicate that SBAs are simple and useful tests for monitoring LT recipients, the prognostic value of SBA levels in the post-LT setting has not been comprehensively investigated. Therefore, this systematic review critically evaluated and synthesized evidence on the association between SBA levels and clinical outcomes after LT.

MATERIALS AND METHODS

Eligibility criteria

Observational studies or clinical trials were included in this systematic review if they adhered to the following criteria: (1) Included adults or pediatric patients undergoing LT; (2) Reported SBA levels measured before and/or after LT; (3) Evaluated the association between SBA levels and post-transplant clinical outcomes, such as acute graft rejection (AGR), graft dysfunction, patient survival, or post-transplant biliary complications; and (4) Were published in English or had an English translation. We excluded case reports, case series with fewer than 5 participants, animal studies, conference abstracts without sufficient data for analysis, review articles, editorials, and letters to the editor were excluded. Studies reporting only conventional biomarkers were also excluded. This review is registered in PROSPERO (CRD42026128743).

Information sources and search strategy

A scoping search for relevant studies published from inception until January 2026 was conducted in PubMed, EMBASE, Web of Science, and the Cumulative Index to Nursing and Allied Health Literature (CINAHL) databases. The search strategy in these databases involved combining free-text terms related to LT, SBA, graft dysfunction, graft rejection, patient survival, and biliary complications. The full search terms are outlined in Appendix A. Additionally, to minimize selection bias, we searched gray literature sources, including ClinicalTrials.gov, the World Health Organization International Clinical Trials Registry Platform, OpenGrey, and ProQuest. Nevertheless, gray literature was not included in the systematic review. We also manually searched the bibliographies of included studies, relevant reviews, and grey literature to identify additional studies that may have been missed during the electronic database search.

Selection process

After completing the database search, the identified studies were imported into the Zotero reference management software and all duplicates were eliminated. Subsequently, two independent authors screened the titles and abstracts of each study and assessed their eligibility using the outlined eligibility criteria. We extracted information on demographics, transplant type, and detection methods; however, these varied across studies. Disagreements during this process were resolved by consulting a third reviewer, who assisted in determining whether a study should be included or excluded. If an abstract was considered eligible, the full study was retrieved. Abstracts that did not have a full manuscript but had sufficient data for review were also included.

Data extraction and data items

Two reviewers independently extracted the data required for review using a standardized Excel spreadsheet. Extracted data included the name of the first author, publication year, study design, country, study population, population characteristics (total sample size, age, and sex distribution), SBA measurement timing, SBA measurement method, outcomes evaluated, and key prognostic findings. Discrepancies in the extracted data were resolved through constructive discussions between the two reviewers or by consulting a third reviewer’s input.

The primary outcomes of this systematic review were graft dysfunction, graft rejection, and patient survival, whereas post-transplant biliary complications were the secondary outcome.

Risk of bias assessment

Two independent reviewers evaluated the risk of bias of the included studies using the Quality in Prognosis Studies (QUIPS) tool. Each study was assessed using this tool based on six domains: Study participation, study attrition, prognostic factor measurement, outcome measurement, study confounding, and statistical analysis and reporting. Several prompt questions were used for each domain, and the risk of bias was judged as low, moderate, or high. Differences were resolved through consensus.

RESULTS

Search results

The electronic search yielded 1296 potential studies, with 788 identified from PubMed, 347 from EMBASE, 92 from Web of Science, and 69 from CINAHL. After removing 754 duplicates, title and abstract screening was performed on 542 studies, of which 527 were excluded. Twelve studies underwent full-text screening using the eligibility criteria outlined earlier, of which 5 studies met the inclusion criteria and were included. The reasons for excluding the remaining 7 studies were as follows: 6 were animal studies, and 1 did not report outcomes of interest (Figure 1).

Figure 1 PRISMA chart for study selection.

Characteristics of included studies

Azer et al[9] included 8 adults who underwent orthotopic liver transplant (OLT). The patients were aged between 16 and 66 years, with only one male and seven females. The 8 patients were consecutively selected and represented different allograft outcomes. Five patients presented with graft dysfunction (two with mild early dysfunction, two with AGR, and one with early mild dysfunction followed by AGR, and three with noncomplicated stable grafts. Individual SBAs [serum concentrations of glycocholic acid (GC) plus glycochenodeoxycholic acid (GCDC) concentrations and the ratio of taurocholic acid (TCA) to taurochenodeoxycholic acid (TCDC)] were measured postoperatively. AGR was diagnosed according to clinical and histological criteria. EAD was characterized by sustained changes in LFT values after transplantation that resolved spontaneously without intervention for AGR, accompanied by liver biopsy findings indicative of nonspecific cholestasis, with or without ischemic alterations.

Janssen et al[10] included 41 patients who underwent OLT. Patients were divided into two groups based on the histological occurrence of acute cellular rejection (ACR): Group I (n = 19) included patients without histologically confirmed ACR, and group II (n = 22) included patients with histologically confirmed ACR. The mean age for group I patients was 50 years (range: 33-66) and for group II patients was 42 years (range: 21-67). Total SBA levels were measured postoperatively using an enzymatic photometrical method, and ACR was defined as a threefold increase in SBA over the individual baseline for 3 consecutive days.

Sundararajan et al[11] included 63 consecutive patients who underwent living donor liver transplant (LDLT). Total SBA levels were measured preoperatively and postoperatively from day 1 to day 14. EAD was defined according to the Olthoffs criteria, and ACR was characterized by a doubling of liver enzymes AST/ALT from baseline, in the absence of other reasons that responded to steroid pulse therapy or increased immunosuppression.

Muraca et al[12] included 20 consecutive patients who underwent OLT. Patients were followed for a mean of 26 days after OLT. Graft rejection was diagnosed based on the criteria outlined by Kemnitz et al[13]. Total SBAs were measured using the enzymatic fluorimetric assay (Sterognost 3-alpha-Flu, Nygaard, Oslo, Norway).

Wu et al[14] examined the usefulness of SBA levels in predicting biliary stricture and survival following LT. This study involved 60 children who underwent LDLT. The average age of these children was 2.04 years, with 24 males and 36 females. Indications for LT included biliary atresia (n = 47), methylmalonic acidemia (n = 5), progressive familial intrahepatic cholestasis (n = 3; 2 with type I, 1 with type III), Alagille syndrome (n = 2), autosomal recessive polycystic kidney disease with congenital liver fibrosis (n = 1), inborn error of bile acid metabolism (n = 1), and Wilson’s disease with fulminant liver failure (n = 1). Total fasting SBA levels were measured postoperatively using an enzymatic assay kit (DZ042A; Diazyme, Poway, CA, United States). Table 1 summarizes the characteristics of the included studies.

Table 1 Study characteristics.

Ref.	Country	Study population	Population characteristics			Timing of the SBA measurement	The SBA measurement method	Reported outcomes
			Sample size	Male/female	Age
Azer et al[9]	Australia	Adult patients who underwent OLT	8	1/7	49.63 years	Daily post-transplant measurements (every 6 hours for the first 2 days and then every 24 hour for the following 12 days)	NR	Graft dysfunction and acute graft rejection
Janssen et al[10]	Germany	Adult patients who underwent OLT	41	NR	Patients without acute rejection: 50 years of age. Patients with acute rejection: 42 years	Daily post-transplant measurements	The enzymatic photometric method	Acute graft rejection
Sundararajan et al[11]	India	Patients who underwent LDLT	63	NR	NR	Preoperative and postoperative data from day 1 to day 14	NR	Graft dysfunction and acute graft rejection
Muraca et al[12]	Germany	Adult patients who underwent OLT	20	NR	NR	Postoperatively	Enzymatic fluorimetric assay method	Graft rejection
Wu et al[14]	Taiwan	Pediatric patients who underwent LDLT	60	24/36	2.04 years	Postoperatively	Enzymatic assay method	Biliary stricture and patient survival

NR: Not reported; OLT: Orthotopic liver transplant; LDLT: Living donor liver transplant; SBA: Serum bile acid.

Risk of bias assessment

Table 2 summarizes the risk of bias findings obtained using the QUIPS tool. Overall, the included studies demonstrated low to moderate heterogeneity across most domains. All studies showed low risk across the study participation and attrition domains. This was because they reported consecutive enrollment of patients and did not report any patients lost to follow-up before graft dysfunction or rejection. The confounding domain demonstrated the highest risk of bias, as only one study adjusted for potential confounders. In addition, most studies demonstrated a moderate risk of bias in the statistical analysis and reporting domain. This was because they did not provide any cutoff values for SBA.

Table 2 Risk of assessment of included studies using the Quality in Prognosis Studies tool.

Ref.	Study participation	Study attrition	Prognostic factor measurement	Outcome measurement	Confounding	Statistical analysis and statistical reporting
Azer et al[9]	Low	Low	Moderate	Low	High	Moderate
Janssen et al[10]	Low	Low	Low	Low	High	Moderate
Sundararajan et al[11]	Low	Low	Moderate	Low	High	Moderate
Muraca et al[12]	Low	Low	Low	Low	High	Moderate
Wu et al[14]	Low	Low	Low	Low	Low	Low

Primary outcomes: Graft dysfunction, graft rejection, and patient survival

Graft dysfunction was assessed in two studies. Azer et al[9] reported that individual SBAs (serum concentrations of GC plus GCDC and the ratio of TCA/TCDC) were remarkably increased in patients with graft dysfunction compared with patients with noncomplicated grafts for 24-48 hours before graft dysfunction was histologically diagnosed. Further analysis revealed that individual SBAs could adequately predict graft dysfunction with 100% sensitivity, specificity, positive predictive value (PPV), and negative predictive value for each. None of the conventional LFTs (bilirubin, albumin, alkaline phosphatase, AST, ALT, and GGT) demonstrated similar predictive values, suggesting that individual SBAs are more sensitive, specific, and reliable than conventional biochemical tests for the early detection of graft dysfunction. Similarly, Sundararajan et al[11] observed higher total SBA levels with an increasing trend in patients with EAD from postoperative days 3 to 14.

Graft rejection was reported in four studies. Azer et al[9] found that individual SBAs could precisely differentiate AGR cases from other causes of dysfunction after OLT. Their results revealed significantly higher concentrations of GC plus GCDC in AGR episodes and lower levels of TCA/TCDC ratio than non-rejection graft dysfunction episodes. Consistent with these findings, Janssen et al[10] reported that the increase in SBA was statistically significant in detecting ACR (P < 0.001). Furthermore, the sensitivity (86%), specificity (100%), and PPV (100%) of SBA in the detection of ACR were relatively high compared with those of conventional biochemical tests (bilirubin and transaminases). Similarly, Sundararajan et al[11] found that the sensitivity and specificity of total SBA in predicting ACR were 94.1% and 61.9%, respectively, with an area under the receiver operating characteristic curve (AUC) of 0.78 (95%CI: 0.653-0.877). Muraca et al[12] also observed an increase in SBAs during graft rejection, further supporting the association between elevated SBAs after LT and graft rejection.

Patient survival was only reported in a cross-sectional study involving 60 pediatric patients who underwent LDLT[14]. This study revealed that the mortality risk after LT for patients with SBA levels greater than 40 μmol/L was significantly higher than for patients with SBA levels ≤ 40 μmol/L (log-rank test, P = 0.004).

Secondary outcome: Post-transplant biliary complications

One study investigated post-transplant biliary complications, specifically biliary stricture. Wu et al[14] observed that an SBA level > 40 μmol/L was the most significant predictor of biliary stricture after LT (odds ratio: 65.65; P = 0.003) after adjusting for potential confounders (gender and GGT). Furthermore, the receiver operating characteristic curve analysis demonstrated that SBA level > 40 μmol/L was predictive of post-transplant biliary stricture (sensitivity: 78%, specificity: 90%, and AUC: 82%; P = 0.002).

DISCUSSION

This systematic review synthesized the available evidence on the prognostic utility of SBA levels after LT, focusing on graft dysfunction, graft rejection, patient survival, and post-transplant biliary complications. Although the available evidence base was heterogeneous, studies consistently showed that elevated SBA levels after LT were associated with adverse clinical outcomes, specifically EAD, AGR, mortality risk, and post-transplant biliary stricture. Variations in demographics, transplant characteristics, and detection methods across different studies may partially account for the inconsistencies in results, and these factors should be considered when applying the findings to particular clinical contexts.

This systematic review suggests that SBA levels can be used to detect EAD after LT and to identify graft dysfunction earlier than conventional biochemical tests. This finding is consistent with previous experimental studies, which have shown that total SBA levels are valuable test for detecting graft dysfunction after LT in animals[7,8,15]. The biological reason for this observation is that bile acid synthesis, uptake, conjugation, and excretion are meticulously controlled hepatic processes that may be swiftly compromised by hepatocellular injury, ischemia-reperfusion damage, and impaired bile flow[16]. Therefore, the disruption of these functions causes rapid accumulation of bile acids in the blood, potentially acting as a biochemical reflection of graft dysfunction. This elevation often occurs before changes in conventional LFTs are observed, which might explain why SBA levels are earlier indicators of graft dysfunction than traditional biochemical tests. Baumgartner et al[17] found that SBA concentration may indicate graft dysfunction 1-3 days earlier than AST concentration.

Evidence across the included studies also showed that SBA levels are sensitive and reliable indicators of AGR after LT, and that SBA levels tend to be elevated during graft rejection episodes. These elevated SBA levels during AGR can be explained by several pathophysiological factors. The immune-mediated attack triggered by lymphocytes on the vascular endothelium and intrahepatic bile ducts of the graft is a key factor. This attack significantly reduces bile acid absorption from the portal circulation and its outflow into the bile ducts. The elevated rate of enterohepatic circulation of bile acids, which occurs approximately 5-15 times daily[18], may also explain the rapid escalation of SBAs in cases of AGR. Another explanation for the increased levels of SBA during graft rejection could be the presence of MRP2. MRP2 is an ATP-regulated conjugate export pump located in the canalicular apical membrane of hepatocytes and is involved in bile acid transport into the bile duct and bilirubin elimination[19,20]. Experimental and clinical studies[21-25] have shown that inflammatory mediators, such as endotoxins and pro-inflammatory cytokines, released by liver-infiltrating lymphocytes during rejection quickly lower MRP2 expression and make it less effective at transporting substances, sometimes in just 16 hours. Therefore, because bile acids circulate through the liver much faster than bilirubin, MRP2 dysfunction has a greater effect on SBA levels. This causes SBA to rise quickly and noticeably, even when bilirubin levels remain the same.

Evidence regarding patient survival is limited, with only one cross-sectional study suggesting that SBA > 40 μmol/L was associated with increased mortality risk[14]. This finding is consistent with that of a previous study, which reported that elevated SBA levels may predict the survival of patients with cirrhosis[26]. Furthermore, SBA > 40 μmol/L was a significant predictor of biliary stricture after LT. This finding was supported by multivariate logistic regression and demonstrated good diagnostic accuracy, suggesting that the SBA level might be a valuable non-invasive method for identifying possible biliary stricture after LT. Nevertheless, due to limited studies and a small sample size of the available evidence, further studies with large cohorts are needed to confirm the relationship between elevated SBA levels and post-transplant biliary complications and patient survival.

Notably, evidence seems to suggest that apart from being used as a prognostic test, SBA levels may be used to monitor antirejection treatment response. Indeed, Azer et al[9] reported that in patients with irreversible graft rejection, the concentration of individual SBAs did not respond to antirejection therapy. However, this was not the case for patients who responded to antirejection treatment, as they demonstrated significant changes in individual SBAs within 24 hours. Similarly, Janssen et al[10] observed a rapid decrease in SBA levels after effective treatment. This was particularly evident in patients with acute rejection, where SBA levels returned to baseline within 3 days after antirejection therapy initiation.

Clinical implications and future research

The findings of this systematic review suggest that SBAs may complement traditional LFTs in the post-liver transplant setting by offering a more sensitive and earlier indicator of graft dysfunction, graft rejection, patient survival, and biliary complications. However, due to the heterogeneity in the methods of SBA assay, timing of SBA measurement, and outcome definitions, further validation in large-sample multicenter studies is required before SBA monitoring can be routinely implemented in clinical practice. Notably, none of the studies investigated the association between SBA levels measured before LT and post-transplant outcomes. This represents a considerable gap in the existing literature and highlights the need for prospective studies with large cohorts to assess the prognostic utility of pre-transplant SBA levels.

Strengths and limitations

The key strength of this systematic review lies in its rigorous methodology, which involved synthesizing all available evidence on the prognostic value of SBA levels after LT, adhering to the PRISMA guidelines, and assessing methodological quality using the QUIPS tool. This rigorous methodology ensured that we provided a broad overview of evidence and strengthened the reliability of our findings. Another strength was the inclusion of human studies only, meaning that our findings directly apply to human health outcomes, thereby increasing the relevance of our findings in clinical and real-world settings.

Nevertheless, this study has limitations. First, the included studies were observational and had small sample sizes, limiting their generalizability. Second, confounding factors were rarely adjusted, increasing the susceptibility to bias. Finally, the heterogeneity in the SBA assay methods, timing of SBA measurement, and outcome definitions precluded quantitative analysis.

CONCLUSION

In summary, the existing evidence suggests that the SBA levels measured after LT can be used as a non-invasive and easily applicable biomarker for monitoring and predicting graft dysfunction, graft rejection, patient survival, and biliary complications. However, further prospective studies are required to confirm the prognostic value of SBA before its routine clinical implementation. Therefore, SBA should be considered a complementary tool rather than a replacement for established biochemical tests.