Factors associated with textbook outcome after robotic-assisted pancreatoduodenectomy: a single surgeon’s retrospective cohort

Introduction

Despite the rapid advancements in surgical techniques, pancreatoduodenectomy (PD) remains one of the most challenging and intricate procedures in the field of general surgery (1). This was exemplified by our recent case, which involved a robotic-assisted pancreaticoduodenectomy (RPD) combined with left hemi-hepatectomy and spleen-preserving resection of the pancreatic tail (2). Considering the intricate nature of PD and its inherent risks, the quality of perioperative management, surgical technique and short-term outcomes may vary considerably between centers (3). Traditionally, the efficacy of PD has been assessed by short-term complications, including postoperative pancreatic fistula (POPF), postpancreatectomy hemorrhage (PPH), delayed gastric emptying (DGE), the length of stay (LOS), mortality, and so on (4,5). However, the single parameters mentioned above are insufficient to comprehensively assess the quality of surgery or to compare the performance across different centers (6,7). Furthermore, understanding the diverse surgical oncological outcomes can be challenging for patients undergoing PD (8), highlighting the urgency for a clear and straightforward metric to assess whether the outcomes have met patients’ optimal expectations.

The concept of textbook outcome (TO) was introduced by the Dutch Colorectal Consortium in 2012 as a composite measurement composed of various postoperative metrics, describing the most desirable outcome after surgery (9). And the definition of TO in the field of pancreatic surgery was proposed by van Roessel et al. in 2020 under an international consensus. TO is achieved when all the predefined outcome parameters are met according to an all-or-none principle, which signifies an ideal postoperative state (10). Previous study has shown that among patients undergoing pancreatic surgery, TO was achieved in 47.8% after partial pancreatectomy like left pancreatectomy and in 27.4% after major pancreatectomy like total pancreatectomy (TP) and PD (11). A multicenter analysis conducted by Wu et al. identified several factors, including dilated pancreatic duct, advanced age and concomitant cardiovascular disease, that were associated with TO in patients after laparoscopic pancreatoduodenectomy (LPD) (7). What’s more, Kwon et al. indicated that patients after RPD had a significantly higher rate of TO than LPD (12). To our knowledge, no studies have yet specifically investigated the TO rate and its associated factors in patients following RPD.

Accordingly, the aim of our study was to evaluate the surgical quality of RPD with TO, and recognize pre- and intraoperative factors that are independently associated with TO. Furthermore, the potential bias in outcomes due to variations in surgical experience among surgeons can be reduced by analyzing the performance of a single surgeon. We present this article in accordance with the STROCSS reporting checklist (available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-24-547/rc).

Methods

Definition

TO in pancreatic surgery is defined by the absence of these 6 parameters: clinically relevant POPF (CR-POPF, grade B/C POPF), bile leak, PPH [both grades B/C according to International Study Group on Pancreatic Surgery (ISGPS)], major complications (Clavien-Dindo ≥ III), readmission within 30 days after discharge, and mortality in-hospital or within 30 days (10).

Study design

This was a retrospective study, which included a 302-patient cohort who underwent RPD by a single surgeon (X.Y.Y.) at The First Affiliated Hospital of Sun Yat-sen University from July 2016 to August 2024. The study was conducted in accordance with the Declaration of Helsinki (and its subsequent amendments). Considering the retrospective nature of this study, informed consent from patients was waived. The ethical approval of the study was obtained from the Ethics Committee of The First Affiliated Hospital of Sun Yat-sen University (approval number: [2024]799).

Surgical procedures and learning curve

The indications for RPD are as follows: (I) resectable pancreatic cancer (RPC) or borderline resectable pancreatic cancer (BRPC); the definition of RPC or BRPC is according to the proposal by National Comprehensive Cancer Network (NCCN) guidelines. For patients with BRPC, neoadjuvant chemotherapy was performed. (II) Periampullary tumors, which include distal cholangiocarcinoma, ampullary carcinoma, and duodenal papillary tumors. (III) Pancreatic neuroendocrine tumors (pNETs) and duodenal neuroendocrine neoplasms (dNENs). (IV) Pancreatic cystic neoplasms (PCNs). (V) Chronic pancreatitis. (VI) Other pancreatic tumors.

All RPD procedures were performed by using the da Vinci Surgical System (Intuitive Surgical, Inc., Sunnyvale, CA, USA). The critical procedures of RPD were briefly described as follows (13).

In the resection part, the major surgical steps included: (I) the Kocher maneuver was performed to isolate the second and third segments of the duodenum and the pancreatic head, ultimately exposing the inferior vena cava, abdominal aorta, left renal vein, and the root of the superior mesenteric artery (SMA); (II) dissecting the superior mesenteric vein (SMV) behind the pancreatic neck, with ligation and division of the right gastroepiploic vessels and Henle’s trunk; (III) transecting the distal stomach with a stapler, then gradually dissecting the common hepatic artery, proper hepatic artery, right gastric artery (RGA), and gastroduodenal artery (GDA), with subsequent division of the RGA and GDA; (IV) transecting the pancreatic neck by the electrocauterization; (V) the proximal jejunum, common bile duct, and uncinate process were divided, along with gallbladder resection.

In the reconstruction part, pancreatojejunostomy (PJ), cholangiojejunostomy (CJ), and gastrojejunostomy (GJ) were briefly described as follows (13,14). For the PJ, the “duct-to-mucosa” anastomosis and other pancreatic anastomosis like the “stent-bridged” anastomosis were performed. The stents were routinely placed across the PJ. For the CJ, either continuous or interrupted suture was performed. Finally, the reconstruction of the GJ was completed using robotic-assisted double-layered suturing.

Our previous research also indicated that the incidence of CR-POPF decreased significantly after the surgeon’s 120th case within the cohort of RPD patients from July 2016 to October 2022 (13).

In this study, we utilized the result and divided 302 patients into two phases: Phase I, the learning phase (cases 1–120), and Phase II, the proficiency phase (cases 121–302).

Data collection

The data were retrospectively and anonymously collected from our in-hospital database.

Preoperative data included age, sex, American Society of Anesthesiologists (ASA) score, comorbidities, admission symptoms, neoadjuvant chemotherapy, preoperative biliary drainage, hemoglobin (Hb), albumin, total bilirubin (TBIL), and carbohydrate antigen 19-9 (CA19-9). Intra- and postoperative variables included duration of operation, estimated blood loss, vascular resection, measurements of the pancreatic and bile ducts, anastomosis techniques, pathological outcome. Short-term outcomes involved LOS, CR-POPF, bile leak, PPH (both grades B/C according to ISGPS), major complications (Clavien-Dindo ≥ III), readmission or mortality within 30 days.

Statistical analysis

TO was achieved in the absence of all the 6 postoperative outcome events, including no CR-POPF, no bile leakage, no PPH, no major complications, no 30-day readmission and no mortality. The frequency distributions of TO, no CR-POPF, no bile leakage, no PPH, no major complications, no 30-day readmission and no mortality were described using histograms. Pre- and intraoperative variables were compared between the non-TO and TO groups. Continuous variables with a normal distribution were described by mean and standard deviation (SD) and analyzed using independent samples t-test. Continuous variables with a non-normal distribution were described by median and interquartile range (IQR) and analyzed using the Mann-Whitney U test. Categorical variables were described by frequencies and analyzed by Chi-square test or Fisher’s exact test. Univariate and multivariate logistic regression analysis was performed to evaluate the variables associated with achieving TO. The results were expressed as odds ratios (OR) with 95% confidence intervals (CI), and variables with a P value <0.1 were included in multivariate regression.

Statistical analysis was performed with R software (version 4.4.1). A P value <0.05 was considered statistically significant.

Results

Distribution of TO and difference analysis

In total, 302 consecutive patients after RPD were enrolled. Among them, 172 (57.0%) achieved TO. Additionally, the TO rate before the surgeon’s learning curve was 47.5%, and in the subsequent cases (n=182) after the learning curve, 115 (63.2%) achieved TO. The frequencies for TO and its six metrics are displayed in Figure 1. The outcome metric least commonly realized was “no Clavien-Dindo grade ≥ III” (67.5%), followed by “no CR-POPF” (79.1%), while the most frequently realized was “no 30-day mortality” (99.0%). A total of 98 patients (32.5%) were classified as Clavien-Dindo ≥ III grade. Among these cases, 58 patients experienced DGE, 50 had intra-abdominal infections, 41 had POPF, 22 had PPH, 13 had cardio-pulmonary-cerebral dysfunction, and 8 had bile leaks. One patient can suffer from more than one postoperative complication.

Figure 1 Distribution of TO and its six metrics in all patients and post-learning curve patients. PPH, postoperative pancreatic hemorrhage; CR-POPF, clinically relevant postoperative pancreatic fistula; TO, textbook outcome.

Among our 302 RPD cases, only five cases were converted to open surgery. One case involved vascular resection due to tumor invasion of the SMV. Two cases were converted due to difficult vascular dissections—one involving the inferior vena cava and abdominal aorta, and the other involving the superior mesenteric artery (SMA). Additionally, one case was combined with a right hemicolectomy, and one case was converted due to anastomotic bleeding that could not be managed robotically.

The baseline characteristics and perioperative data between TO and non-TO groups are presented in Table 1. Significant differences in ASA score, surgeon’s learning curve, concomitant diabetes, and pancreatic anastomosis were shown between TO and non-TO groups (all P<0.05). The patients in TO group had a wider diameter of pancreatic duct (Z=−3.38, P<0.001) and larger tumor size (Z=−2.11, P=0.04). Notably, the median LOS in the TO group was 12 days, which is significantly shorter than that in the non-TO group (Z=−10.64, P<0.001). There was no significant difference in other data presented in Table 1 between the two groups (all P>0.05).

Factors associated with textbook outcome

On our multivariate logistic regression analysis as shown in Table 2, cases after learning curve (OR: 2.54, 95% CI: 1.39–4.64; P=0.002), concomitant diabetes (OR: 3.25, 95% CI: 1.33–7.93; P=0.009), dilated pancreatic duct (>3 mm) (OR: 1.90, 95% CI: 1.14–3.18; P=0.01), and ASA score ≤ II (OR: 0.34, 95% CI: 0.19–0.60; P<0.001) were independently associated with achieving TO after RPD.

Discussion

To our knowledge, this is the first study to use TO to evaluate the quality in patients undergoing RPD and to explore perioperative factors that are associated with TO.

Adopting the definition of TO in pancreatic surgery as previously mentioned (10), TO was achieved in 57.0% of all patients undergoing RPD and 63.2% of patients undergoing RPD after the learning curve of a surgeon. The major obstacle preventing the occurrence of TO is the metric of “no Clavien-Dindo grade ≥ III” compared with the other five metrics.

In this study, several preoperative and operative factors were found to be independently associated with the achievement of TO after RPD, including surgeon’s learning curve, ASA score, concomitant diabetes, dilated pancreatic duct (>3 mm) or not.

Previous studies have shown that as the learning curve phases advance, the incidences of major complications (Clavien-Dindo ≥ III) and POPF after PD gradually decrease (13,15). In our multivariate analysis, we found that the surgeon’s learning curve was an independent predictive factor for achieving TO, which was consistent with previous studies. The reasons could be that the surgeon’s learning curve helped to cut down major complications (Clavien-Dindo ≥ III) and POPF, which were the criteria of excluding TO in pancreatic surgery (10).

According to existing studies in general surgery, patients with ASA grade ≥ III were recognized as a risk factor for achieving TO in other surgeries, such as gastrectomy (16). A previous study has shown that patients with ASA grade ≥ III had a negative impact on achieving ideal outcome after PD, which was a novel definition derived from TO in pancreatic surgery (6). Our study showed that patients with ASA grade ≥ III had a decreased probability of achieving TO, probably because they may have experienced a reduced capacity to the stress of general anesthesia, which may lead to the higher incidence rate of perioperative cardiopulmonary complications (17). Besides, a previous study indicated that patients with ASA ≥ III grade exhibited a reduced ability to cope with complications after PD, leading to elevated in-hospital mortality rates and a higher failure-to-rescue rate (18).

Interestingly, we found that concomitant diabetes was a protective factor for achieving TO in pancreatic surgery, a finding that may not be consistent with the general belief. An increased risk of postoperative complications in patients with preoperative diabetes has been reported in other fields (19). Several studies have suggested that diabetes may lead to an increased risk of poor short-term prognosis in patients after pancreatic surgery (20,21). However, Eshuis et al. reported that preoperative glucose values were not associated with postoperative complications (22). Notably, preoperative diabetes was associated with a lower incidence of clinically significant POPF, probably due to a lower frequency of soft pancreatic texture resulting from reduced pancreatic fat (23). This condition has been identified by the ISGPS as a risk factor for POPF after pancreatic surgery (24). Investigations into the relationship between preoperative diabetes and TO after RPD are rare. Although our study suggests that diabetes is an independent protective factor for TO after RPD, further study is needed to clarify the relationship between diabetes and TO.

In our study, dilated pancreatic duct (>3 mm) has been identified as a protective factor for achieving TO after RPD. Previous studies and consensus by the ISGPS proved that dilated pancreatic duct (>3 mm) was strongly associated with lower incidence of POPF (24,25), and our finding is consistent with their results. Besides, a majority of randomized controlled trials showed that the impact of duct-to-mucosa pancreatic anastomosis compared to other pancreatic anastomosis techniques on CR-POPF, postoperative mortality, and overall rate of surgical complications is very uncertain (26,27). Our study suggests that the anastomosis technique is not associated with achieving TO, possibly due to the similar occurrence of POPF across both techniques.

Above all, recognizing these independent factors preoperatively is important to surgeons, as early and preoperative identification allows for more attention and management strategies that may enhance the likelihood of achieving TO after RPD. Previous studies have indicated that factors associated with TO after LPD include age, sex, dilated pancreatic duct (>3 mm), pancreatic texture, and concomitant cardiovascular disease, which are mostly different from the factors we analyzed in our article (7,28). The existing studies mainly focus on comparing the TO rates between different surgical approaches of PD [for example, a previous literature suggested that patients after RPD had a significantly higher rate of TO than LPD (12)]. Based on the current literature, there has been a lack of comparative studies regarding the factors associated with achieving TO between open pancreatoduodenectomy (OPD) and RPD, or between LPD and RPD. Further comparative study is needed. Notably, we found that the median LOS in the TO group was a mere 12 days, significantly shorter than that of the non-TO group. This reduction in LOS is particularly beneficial for patients with malignant tumors, as it allows for the earlier initiation of adjuvant therapy, thus improving their long-term prognosis. In our study, achieving TO signifies not only an enhancement in the quality of surgery but also a reduction in the postoperative hospital stay for patients. It is also worth mentioning that in other fields, LOS is considered an integral component of the TO definition, which may contribute to a reduction in the TO rate (29-31).

The limitations of our study are as follows. Firstly, this is a retrospective study with its inherent biases. Secondly, due to a lack of adequate cases of LPD in our center, we did not compare the TO rate and perioperative parameters between LPD and RPD. Thirdly, while the definition of TO in some other surgical fields includes LOS, our study adopted the Rossel definition of TO for PD due to the following reasons. In China, the majority of PD procedures are performed in the major metropolitan medical centers, like ours. Many patients from the suburban areas or other small cities are reluctant to be discharged from the medical centers after PD, even meeting the discharge criteria. It would lead to some unnecessary prolongation of LOS, and inclusion of LOS would lead to the bias in real TO. Hence, we believe that the Rossel definition for pancreatic surgery could reflect the real TO in PD.

Conclusions

TO was achieved in more than half of the patients after RPD. Surgeon’s learning curve, ASA score, concomitant diabetes, and dilated pancreatic duct (>3 mm) were independently associated with TO.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROCSS reporting checklist. Available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-24-547/rc

Data Sharing Statement: Available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-24-547/dss

Peer Review File: Available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-24-547/prf

Funding: This study was supported by the National Natural Science Foundation of China (No. 82203105), Guangzhou Science and Technology Plan Project (No. 2023A04J2211), and The Guangzhou Clinical Major Technical Project (2023P-ZD02).

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://hbsn.amegroups.com/article/view/10.21037/hbsn-24-547/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (and its subsequent amendments). Considering the retrospective nature of this study, informed consent from patients was waived. The ethical approval of the study was obtained from the Ethics Committee of The First Affiliated Hospital of Sun Yat-sen University (approval number: [2024]799).

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Hays SB, Rojas AE, Hogg ME. Robotic pancreas surgery for pancreatic cancer. Int J Surg 2024;110:6100-10. [Crossref] [PubMed]
2. Huang XT, Cai JP, Yin XY. Robotic-assisted hepatopancreatoduodenectomy and resection of pancreatic tail for one multiple endocrine neoplasia type 1 secreting multiple hormones. Hepatobiliary Surg Nutr 2023;12:818-21. [Crossref] [PubMed]
3. Macedo FIB, Jayanthi P, Mowzoon M, et al. The Impact of Surgeon Volume on Outcomes After Pancreaticoduodenectomy: a Meta-analysis. J Gastrointest Surg 2017;21:1723-31. [Crossref] [PubMed]
4. Boone BA, Zenati M, Hogg ME, et al. Assessment of quality outcomes for robotic pancreaticoduodenectomy: identification of the learning curve. JAMA Surg 2015;150:416-22. [Crossref] [PubMed]
5. Shi Y, Jin J, Qiu W, et al. Short-term Outcomes After Robot-Assisted vs Open Pancreaticoduodenectomy After the Learning Curve. JAMA Surg 2020;155:389-94. [Crossref] [PubMed]
6. Augustinus S, Mackay TM, Andersson B, et al. Ideal Outcome After Pancreatoduodenectomy: A Transatlantic Evaluation of a Harmonized Composite Outcome Measure. Ann Surg 2023;278:740-7. [Crossref] [PubMed]
7. Wu Y, Peng B, Liu J, et al. Textbook outcome as a composite outcome measure in laparoscopic pancreaticoduodenectomy: a multicenter retrospective cohort study. Int J Surg 2023;109:374-82. [Crossref] [PubMed]
8. Liu ZP, Guo W, Yin DL, et al. Textbook outcomes in liver surgery for gallbladder cancer patients treated with curative-intent resection: a multicenter observational study. Int J Surg 2023;109:2751-61. [Crossref] [PubMed]
9. Kolfschoten NE, Kievit J, Gooiker GA, et al. Focusing on desired outcomes of care after colon cancer resections; hospital variations in 'textbook outcome'. Eur J Surg Oncol 2013;39:156-63. [Crossref] [PubMed]
10. van Roessel S, Mackay TM, van Dieren S, et al. Textbook Outcome: Nationwide Analysis of a Novel Quality Measure in Pancreatic Surgery. Ann Surg 2020;271:155-62. [Crossref] [PubMed]
11. Merath K, Chen Q, Bagante F, et al. Textbook Outcomes Among Medicare Patients Undergoing Hepatopancreatic Surgery. Ann Surg 2020;271:1116-23. [Crossref] [PubMed]
12. Kwon J, Kang CM, Jang JY, et al. Perioperative textbook outcomes of minimally invasive pancreatoduodenectomy: a multicenter retrospective cohort analysis in a Korean minimally invasive pancreatic surgery registry. Int J Surg 2024;110:4249-58. [Crossref] [PubMed]
13. Huang XT, Wang XY, Xie JZ, et al. Learning curves of resection and reconstruction procedures in robotic-assisted pancreatoduodenectomy by a single surgeon: a retrospective cohort study of 160 consecutive cases. Gastroenterol Rep (Oxf) 2023;11:goad042. [Crossref] [PubMed]
14. Huang XT, Qu XF, Zhou JW, et al. Robotic-assisted organ-preserving or parenchymal-sparing pancreatectomy in pancreatic benign or low-grade malignant tumors: a single institute's experience. J Robot Surg 2024;18:1.
15. Peng F, He R, Wang H, et al. Development of a difficulty scoring system for laparoscopic pancreatoduodenectomy in the initial stage of the learning curve: a retrospective cohort study. Int J Surg 2023;109:660-9. [Crossref] [PubMed]
16. Sędłak K, Rawicz-Pruszyński K, Mlak R, et al. Textbook Oncological Outcome in European GASTRODATA. Ann Surg 2023;278:823-31. [Crossref] [PubMed]
17. Hightower CE, Riedel BJ, Feig BW, et al. A pilot study evaluating predictors of postoperative outcomes after major abdominal surgery: Physiological capacity compared with the ASA physical status classification system. Br J Anaesth 2010;104:465-71. [Crossref] [PubMed]
18. Sánchez-Velázquez P, Muller X, Malleo G, et al. Benchmarks in Pancreatic Surgery: A Novel Tool for Unbiased Outcome Comparisons. Ann Surg 2019;270:211-8. [Crossref] [PubMed]
19. Tan DJH, Yaow CYL, Mok HT, et al. The influence of diabetes on postoperative complications following colorectal surgery. Tech Coloproctol 2021;25:267-78. [Crossref] [PubMed]
20. Eshuis WJ, Hermanides J, van Dalen JW, et al. Early postoperative hyperglycemia is associated with postoperative complications after pancreatoduodenectomy. Ann Surg 2011;253:739-44. [Crossref] [PubMed]
21. Zhang X, Sun C, Fei H, et al. Impact of Diabetes on Short-Term and Long-Term Outcomes of Ampullary Adenocarcinoma Patients after Curative Pancreatoduodenectomy. Curr Oncol 2022;29:6724-34. [Crossref] [PubMed]
22. Nakata B, Ishikawa T, Amano R, et al. Impact of preoperative diabetes mellitus on clinical outcome after pancreatectomy. Int J Surg 2013;11:757-61. [Crossref] [PubMed]
23. Xia X, Huang C, Cen G, et al. Preoperative diabetes as a protective factor for pancreatic fistula after pancreaticoduodenectomy: a meta-analysis. Hepatobiliary Pancreat Dis Int 2015;14:132-8. [Crossref] [PubMed]
24. Schuh F, Mihaljevic AL, Probst P, et al. A Simple Classification of Pancreatic Duct Size and Texture Predicts Postoperative Pancreatic Fistula: A classification of the International Study Group of Pancreatic Surgery. Ann Surg 2023;277:e597-608. [Crossref] [PubMed]
25. Huang XT, Huang CS, Liu C, et al. Development and Validation of a New Nomogram for Predicting Clinically Relevant Postoperative Pancreatic Fistula After Pancreatoduodenectomy. World J Surg 2021;45:261-9. [Crossref] [PubMed]
26. Wang W, Zhang Z, Gu C, et al. The optimal choice for pancreatic anastomosis after pancreaticoduodenectomy: A network meta-analysis of randomized control trials. Int J Surg 2018;57:111-6. [Crossref] [PubMed]
27. Hai H, Li Z, Zhang Z, et al. Duct-to-mucosa versus other types of pancreaticojejunostomy for the prevention of postoperative pancreatic fistula following pancreaticoduodenectomy. Cochrane Database Syst Rev 2022;3:CD013462. [Crossref] [PubMed]
28. Cai H, Lu F, Gao P, et al. Risk factors of textbook outcome in laparoscopic pancreatoduodenectomy: results from a prospective high-volume center study. BMC Surg 2024;24:233. [Crossref] [PubMed]
29. Azevedo JM, Panteleimonitis S, Mišković D, et al. Textbook Oncological Outcomes for Robotic Colorectal Cancer Resections: An Observational Study of Five Robotic Colorectal Units. Cancers (Basel) 2023;15:3760. [Crossref] [PubMed]
30. Melgar P, Rodríguez-Laiz GP, Lluís N, et al. Textbook outcome among patients undergoing enhanced recovery after liver transplantation stratified by risk. A single-center retrospective observational cohort study. Int J Surg 2022;99:106266. [Crossref] [PubMed]
31. Xu Z, Lv Y, Zou H, et al. Textbook outcome of laparoscopic hepatectomy in the context of precision surgery: A single center experience. Dig Liver Dis 2024;56:1368-74. [Crossref] [PubMed]