Abstract
Background
Posthepatectomy liver failure remains a potentially life-threatening complication after hepatectomy. Soluble suppression of tumourigenicity 2 is an injury-related biomarker. The aim of the study was to assess soluble suppression of tumourigenicity 2 elevation after hepatectomy and whether it can predict posthepatectomy liver failure.
Methods
This was a single-centre retrospective study including all patients who underwent a liver resection between 2015 and 2019. Plasma concentrations of soluble suppression of tumourigenicity 2 were measured before surgery and at postoperative days 1, 2, 5 and 7. Posthepatectomy liver failure was defined according to the International Study Group of Liver Surgery and the morbidity rate was graded according to the Clavien–Dindo classification.
Results
A total of 173 patients were included (75 underwent major and 98 minor resection); plasma levels of soluble suppression of tumourigenicity 2 increased from 43.42 (range 18.69–119.96) pg/ml to 2622.23 (range 1354.18–4178.27) pg/ml on postoperative day 1 (P < 0.001). Postoperative day 1 soluble suppression of tumourigenicity 2 concentration accurately predicted posthepatectomy liver failure ≥ grade B (area under curve = 0.916, P < 0.001) and its outstanding performance was not affected by underlying disease, liver pathological status and extent of resection. The cut-off value, sensitivity, specificity, positive predictive value and negative predictive value of postoperative day 1 soluble suppression of tumourigenicity 2 in predicting posthepatectomy liver failure ≥ grade B were 3700, 92%, 85%, 64% and 97% respectively. Soluble suppression of tumourigenicity 2high patients more frequently experienced posthepatectomy liver failure ≥ grade B (64.3% (n = 36) versus 2.6% (n = 3)) and Clavien–Dindo IIIa higher morbidity rate (23.2% (n = 13) versus 5.1% (n = 6)) compared with soluble suppression of tumourigenicity 2low patients.
Conclusions
Soluble suppression of tumourigenicity 2 may be a reliable predictor of posthepatectomy liver failure ≥ grade B as early as postoperative day 1 for patients undergoing liver resection. Its role in controlling hepatic injury/regeneration needs further investigation.
Registration number: ChiCTR-OOC-15007210 (www.chictr.org.cn/).
This study proved the power of soluble suppression of tumourigenicity 2 in predicting posthepatectomy liver failure in order to start early intervention.
Introduction
When performed in an experienced centre, hepatic resections have good postoperative outcomes1. Posthepatectomy liver failure (PHLF), however, is still the most common cause of postoperative death2.
Preoperative scoring systems3,4, three-dimensional reconstructions5 and preoperative interventions6 have reduced the PHLF rate, but it remains the most lethal complication, especially after extended hepatectomy, also in non-cirrhotic patients7. To predict PHLF within the first 24 h after hepatectomy would be of paramount importance for initiating interventions to treat hepatic dysfunction8 and/or start the difficult process of emergency liver transplantation9.
Nine circulating markers potentially related to liver failure/regeneration were measured in a preliminary study including 40 patients who underwent hepatectomy, with interleukin 33 (IL-33) and suppression of tumourigenicity 2 (ST2) being the most accurate. IL-33, described as a cellular alarmin, is released from damaged tissue-resident cells, binds to the ST2 receptor and induces an inflammatory response10.
Indeed, the blood concentration of soluble ST2 (sST2) increases in several inflammatory diseases11,12. sST2 is currently emerging as a useful prognostic marker in patients with heart failure13,14 and is considered a promising marker for the treatment of liver diseases15–17. The aim of the study was to assess sST2 elevation after hepatectomy and whether it can predict PHLF.
Methods
Ethical statement
This single-centre retrospective study included all patients who underwent hepatic resection between 2015 and 2019 (Fig. S1). The study was conducted according to the Helsinki Declaration. It was registered at the Chinese Clinical Trial Registry Center (ChiCTR-OOC-15007210), approved by the Institutional Ethical Committee (20151010–2) and reported in line with the Standards for Reporting of Diagnostic Accuracy Studies (STARD) criteria18. 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. This study was performed in accordance with the Helsinki Declaration as revised in 2013 and was approved by the Institutional Ethical Committee of the first affiliated hospital of Xinjiang Medical University (20151010-2).
Inclusion and exclusion criteria
Patients were included if they were older than 18 years and underwent hepatic resection. Exclusion criteria were autoimmune disease, chronic consumption of a non-steroid anti-inflammatory drug, hormonal therapy or morphine.
Perioperative strategies
The presence of liver steatosis was assessed by preoperative ultrasonography and graded as mild, moderate and severe19. For all patients, Child–Pugh, model for end-stage liver diseases (MELD)3 and albumin-bilirubin (ALBI) scores were calculated20. The type of resection was graded as major (three or more segments) or minor (fewer than three segments) according to the International Hepato-Pancreato-Biliary Association (IHPBA) Brisbane 2000 nomenclature21. PHLF was defined according to the International Study Group of Liver Surgery (ISGLS) (Table S1)22. Intraoperative parameters including operating time, blood loss, blood transfusion, portal clamping duration and postoperative complications were recorded.
Plasma and tissue analysis
Blood samples were collected before surgery and at PODs 1, 2, 5 and 7. Plasma levels of sST2 and IL-33 were assessed by enzyme-linked immunosorbent assay (CUSABIO, China). The resected specimen was used to assess cirrhotic status and to perform immunostaining for IL-33 and ST2 (see the Supplementary material for detailed information).
Statistical analysis
Results were described in terms of number and proportion (n, %) for qualitative data and medians and interquartile range (i.q.r.) or means and standard deviations (s.d.) for quantitative data. Variables were compared using the Students’ t test, Pearson correlation test, or χ2 test as appropriate. A receiver operating characteristic (ROC) curve analysis was carried out to plot the specificity and sensitivity of circulating sST2 levels in predicting PHLF. The optimal cut-off was chosen by using the maximized Youden index. All parameters with P < 0.100 in univariable analysis were included in a multivariable analysis. Binary logistic regression was performed. P < 0.05 was considered statistically significant. The sample size estimate was performed using PASS version 11.0 and parameters were set as follows. The prevalence of PHLF was estimated around 30%, and the significance level was set to 0.05 with 90% power (1-β). The alternative sensitivity and specificity were defined as 0.90. Statistical analyses were carried out using SPSS version 22.0 (IBM, Armonk, New York, USA).
Results
Baseline information
Baseline information and clinical characteristics of patients in the whole cohort are summarized in Table 1, and those of patients with PHLF are described in Table 2. The clinical characteristics of patients who underwent major and minor hepatectomies are shown in Table S2. Operating time, total/postoperative duration of hospital stay, intraoperative blood loss, blood product transfusion volumes and total portal clamping time were significantly higher in patients with major resection (P < 0.001; Table S2). During the study interval, only two patients were operated on for intrahepatic cholangiocarcinoma and three for colorectal cancer liver metastases, but they were not enrolled due to patient refusal.
Table 1.
Patients’ characteristics of the study population
| Items | Whole cohort (n = 173) |
|---|---|
| Age (years), median (i.q.r.) | 44.50 (34.25–55.00) |
| Sex | |
| Male | 92 |
| Female | 81 |
| BMI (kg/m2), mean(s.d.) | 23.36(3.81) |
| Tumour type | |
| HCC | 47 (27.17) |
| HAE | 47 (27.17) |
| HCE | 35 (20.23) |
| HEM | 24 (13.87) |
| Others | 20 (11.56) |
| Child–Pugh classification | |
| Grade A | 126 (72.83) |
| Grade B | 47 (27.17) |
| ALBI score | |
| Grade 1 | 105 (60.69) |
| Grade 2 | 67 (38.73) |
| Grade 3 | 1 (0.58) |
| MELD score, median (i.q.r.) | 7.19 (6.65–7.90) |
| Fatty liver | |
| None | 146 (84.39) |
| Mild | 21 (12.14) |
| Moderate | 5 (2.89) |
| Severe | 1 (0.58) |
| HBV infection | 37 (21.39) |
| Previous treatment | |
| Chemotherapy | 16 (9.25) |
| Abdominal surgery | 52 (30.06) |
| PVE | 4 (2.31) |
| TACE | 5 (2.89) |
| Smoker | 33 (19.08) |
Values are n (%) unless otherwise stated. i.q.r., interquartile range; HCC, hepatocellular carcinoma; HAE, hepatic alveolar echinococcosis; HCE, hepatic cystic echinococcosis; HEM, haemangioma; ALBI, albumin-bilirubin; MELD, model for end-stage liver diseases; HBV, hepatic B virus; PVE, portal vein embolization; TACE, trans-arterial chemoembolization.
Table 2.
Characteristics of the patients with and without PHLF ≥ grade B
| Items | PHLF < grade B (n = 134) |
PHLF ≥ grade B (n = 39) |
P |
|---|---|---|---|
| Age (years), mean(s.d.) | 44.30(15.68) | 45.03(14.97) | 0.798 |
| Sex | |||
| Male | 71 (52.99) | 21 (53.85) | 0.536 |
| Female | 63 (47.01) | 18 (46.15) | |
| BMI (kg/m2), median (i.q.r.) | 23 (21–26) | 23 (21–27) | 0.967 |
| Type of tumour | 0.000 | ||
| HCC | 32 (23.88) | 15 (38.46) | 0.072 |
| HAE | 26 (19.40) | 21 (53.85) | 0.000 |
| HCE | 35 (26.12) | 0 (0) | 0.000 |
| HEM | 22 (26.42) | 2 (5.13) | 0.073 |
| Others | 19 (14.18) | 1 (2.56) | 0.087 |
| Child–Pugh classification | 0.003 | ||
| Grade A | 105 (78.36) | 21 (53.85) | |
| Grade B | 29 (21.64) | 18 (46.15) | |
| ALBI score | 0.015 | ||
| Grade 1 | 88 (65.67) | 17 (43.59) | |
| Grade 2 | 45 (33.58) | 22 (56.41) | |
| Grade 3 | 1 (0.75) | 0 (0) | |
| MELD score, median (i.q.r.) | 7.19 (6.65–7.83) | 7.00 (6.62–8.04) | 0.610 |
| Fatty liver | 1.000 | ||
| None | 113 (84.33) | 33 (84.62) | |
| Mild | 16 (11.94) | 5 (12.82) | |
| Moderate | 4 (2.99) | 1 (2.56) | |
| Severe | 1 (0.75) | 0 (0) | |
| HBV* | 24 (17.91) | 14 (35.90) | 0.027 |
| Previous treatment | |||
| Chemotherapy | 11 (8.21) | 5 (12.82) | 0.361 |
| Abdominal surgery | 43 (32.09) | 9 (23.08) | 0.326 |
| PVE | 1 (0.75) | 3 (7.69) | 0.036 |
| TACE | 4 (2.99) | 1 (2.56) | 1.000 |
| Allergy | 11 (8.21) | 1 (2.56) | 0.303 |
| Smoker | 25 (18.66) | 8 (20.51) | 0.818 |
| Surgical resection | 0.000 | ||
| Major resection | 42 (31.34) | 33 (84.62) | |
| Minor resection | 92 (68.66) | 6 (15.38) | |
| Operative time (min), median (i.q.r.) | 210 (165–325) | 590 (300–945) | 0.000 |
| Duration of hospital stay (days), median (i.q.r.) | 15 (11–19) | 31 (21–42) | 0.000 |
| Postoperative hospital stay (days), median (i.q.r.) | 7 (6–11) | 19 (11–27) | 0.000 |
| Blood loss (ml), median (i.q.r.) | 250 (100–525) | 1000 (600–1500) | 0.000 |
| Blood transfusion, median (i.q.r.) | |||
| RBC (UI) | 0 (0–0) | 4 (0.50–8.38) | 0.000 |
| Fresh frozen plasma (ml) | 0 (0–0) | 500 (0–820) | 0.000 |
| Autotransfusion (ml) | 0 (0–0) | 0 (0–921) | 0.000 |
| Portal clamping (min), median (i.q.r.) | 0 (0–15) | 70 (18.75–108.00) | 0.000 |
| Postoperative complication | |||
| IIIa or higher | 6 (4.45) | 13 (33.33) | 0.000 |
| Death† | 0 (0) | 4 (10.25) | 0.002 |
Values are n (%) unless otherwise stated. i.q.r., interquartile range; PHLF, posthepatectomy liver failure; HCC, hepatic cellular cancer; HAE, hepatic alveolar echinococcosis; HCE, hepatic cystic echinococcosis; HEM, haemangioma; ALBI, albumin-bilirubin; MELD, model for end-stage liver diseases; HBV, hepatitis B virus; HCV, hepatitis C virus; PVE, portal vein embolization; TACE, trans-arterial chemoembolization; PTBD, percutaneous transhepatic biliary drainage; RBC, red blood cell. *There was one patient with HCV infection in the whole cohort. †Death was included in grade IIIa or higher complication.
Plasma sST2 levels
In patients without PHLF, plasma levels of IL-33 (92.65 (range 48.44–157.54) pg/ml before surgery versus 178.64 (range 110.72–359.40) pg/ml POD1; P < 0.05) and sST2 (43.42 (range 18.69–119.96) pg/ml before surgery versus 2622.23 (range 1354.18–4178.27) pg/ml POD1; P < 0.001) significantly increased on POD1 compared with the preoperative interval. They decreased until POD7 but remained higher than the baseline value (Fig. 1a and Fig. S2a). Circulating levels of sST2 were higher in patients undergoing major versus minor liver resection (Fig. 1b) on PODs 1, 2 and 5 (P < 0.01 respectively). In patients with PHLF, sST2 levels remained significantly higher on POD7 (Fig. 1c, P < 0.05). Levels of sST2 on POD1 in patients without PHLF were significantly inferior to those with different grades of PHLF (Fig. 1d, P < 0.05).
Fig. 1.
The dynamic changes of plasma POD1-sST2 in different hepatic backgrounds
Circulating sST2 levels in a, the whole cohort, in b, patients undergoing minor and major resection, in c, patients with/without PHLF, and in d, patients with different grades of PHLF at predetermined time points. POD, postoperative day; sST2, soluble suppression of tumourigenicity 2; *P < 0.050; **P < 0.010; ***P < 0.001.
Subgroup analysis of plasma IL-33 and sST2 levels
sST2 levels on POD1 were significantly increased in patients with hepatic alveolar echinococcosis (HAE) and hepatocellular carcinoma (HCC) compared with those with hepatic cystic echinococcosis (HCE) (P < 0.05, P < 0.05 respectively) (Fig. 2a, b). sST2 levels on POD1 were significantly higher in hepatitis B virus (HBV) infected patients (Fig. 2c). The association between perioperative sST2 levels and PHLF was evaluated separately for each tumour type (Table S3). Plasma IL-33 levels showed no significant trends in subgroup analyses regarding the resection extent and PHLF grade (Fig. S2b–f). Preoperative IL-33 and sST2 plasma levels were not significantly different in patients with or without cirrhosis and steatosis of different severity (Fig. 2d, e and Fig. S3a, b); plasma IL-33 and sST2 levels as well as IL-33 and ST2 liver expression levels were higher in patients with HCC and HAE (Fig. S3c–e).
Fig. 2.
Impact of underlying liver disease on circulating POD1-sST2 level
a, b, POD1-sST2 levels in patients with hepatocellular carcinoma (HCC), hepatic alveolar echinococcosis (HAE), hepatic cystic echinococcosis (HCE) and haemangioma (HEM). c, Association between POD1 sST2 and hepatitis B virus (HBV) infection. Perioperative comparison of plasma sST2 levels in d, patients with or without steatosis, e, patients with or without cirrhosis. POD, postoperative day; PHLF, posthepatectomy liver failure; sST2, soluble suppression of tumourigenicity 2. n.s., no significance. *P < 0.050; **P < 0.010; ***P < 0.001.
Univariable and multivariable analysis of PHLF
HBV infection, tumour type, ascites, Child–Pugh score, ALBI score, type of resection, blood loss, preoperative parameters (haemoglobin (Hb), aspartate transaminase (AST) and alanine transaminase (ALT)) levels and POD1 parameters (AST, ALT and sST2) predicted PHLF in univariable analysis (all P < 0.1). Blood loss, type of resection and sST2 on POD1 were independent risk factors in multivariable analysis (all P < 0.05) (Table 3).
Table 3.
Univariable and multivariable logistic regression analyses for PHLF ≥ grade B
| Items | Univariable analysis | Multivariable analysis | ||
|---|---|---|---|---|
| OR | 95% c.i. | OR | 95% c.i. | |
| Sex | 0.966 | 0.472,1.975 | ||
| Age (years) | 1.003 | 0.980,1.027 | ||
| Previous surgery | 0.660 | 0.296,1.472 | ||
| Smoking | 1.125 | 0.462,2.741 | ||
| HBV infection | 2.567 | 1.166,5.652 | 1.273 | 0.130,12.457 |
| Tumour type | ||||
| Echinococcosis | ||||
| HCC | 1.362 | 0.619,2.996 | 1.068 | 0.067,16.953 |
| HEM | 0.264 | 0.057,1.220 | 0.422 | 0.037,4.74 |
| Others | 0.153 | 0.019,1.213 | 7.846 | 0.111,556.953 |
| Fatty liver | 0.913 | 0.436,1.913 | ||
| Ascites | 2.140 | 0.963,4.756 | 2.337 | 0.223,24.514 |
| Child–Pugh score | 2.935 | 1.484,5.805 | 0.967 | 0.147,6.365 |
| ALBI score | 2.278 | 1.130,4.594 | 0.520 | 0.096,2.811 |
| Type of resection | 0.085 | 0.033,0.218 | 0.150 | 0.024,0.950 |
| Blood loss (ml) | 1.001 | 1.001,1.002 | 1.002 | 1.000,1.003 |
| Preop. parameters | ||||
| NE percentage (%) | 0.982 | 0.919,1.050 | ||
| NE count | 0.884 | 0.582,1.343 | ||
| Hb (g/l) | 0.982 | 0.964,0.999 | 0.972 | 0.932,1.013 |
| PLT (109/l) | 0.999 | 0.995,1.003 | ||
| AST (UI/l) | 1.010 | 1.001,1.019 | 1.040 | 1.000,1.082 |
| ALT (UI/l) | 1.005 | 0.999,1.011 | 0.976 | 0.949,1.005 |
| IL-33 (pg/ml) | 0.999 | 0.996,1.002 | ||
| sST2 (pg/ml) | 1.002 | 0.999,1.005 | ||
| POD1 parameters | ||||
| NE per cent (%) | 1.008 | 0.937,1.085 | ||
| NE count | 0.978 | 0.858,1.115 | ||
| AST (UI/l) | 1.001 | 1.000,1.002 | 0.998 | 0.996,1.001 |
| ALT (UI/l) | 1.002 | 1.001,1.003 | 1.002 | 0.998,1.006 |
| IL-33 (pg/ml) | 1.001 | 1.000,1.002 | ||
| sST2 (pg/ml) | 1.001 | 1.001,1.002 | 1.002 | 1.001,1.003 |
HBV, hepatitis B virus; OR, odds ratio; HCC, hepatic cellular cancer; HAE, hepatic alveolar echinococcosis; HCE, hepatic cystic echinococcosis; HEM, haemangioma; ALBI, albumin-bilirubin; Hb, haemoglobin; PLT, platelet; ALT, alanine transaminase; AST, aspartate transaminase; IL-33, interleukin 33; sST2, soluble suppression of tumourigenicity 2; POD, postoperative day; NE, neutrophil. The variables indicated in bold are statistically significant.
Plasma sST2 levels and clinical outcome
The area under curve (AUC) of sST2 levels on POD1 for PHLF ≥ grade B in the whole cohort and in the major resection cohort was 0.916 and 0.887 respectively, IL-33 was 0.587 (Fig. 3a, b and Fig. S2e, f). Among patients who underwent major liver resection, sST2 levels on POD1 discriminated well between PHLF ≥ grade B and PHLF < grade B (AUC = 0.8874, P < 0.001) regardless of tumour type and underlying liver disease (Fig. 3c, d).
Fig. 3.
Receiver operating characteristic (ROC) curve analysis of circulating POD1-sST2 for predicting PHLF ≥ grade B
The ROC curve of multi-index in predicting PHLF ≥ grade B in a, the whole cohort and b, the major resection cohort. The predictive power of plasma POD1-sST2 discriminating PHLF ≥ grade B in c, patients with different types of disease, and d, patients with or without hepatitis B virus (HBV) infection. POD, postoperative day; AUC, area under curve; ROC, receiver operating characteristic; sST2, soluble suppression of tumourigenicity 2; PHLF, posthepatectomy liver failure; ALBI, albumin-bilirubin; HCC, hepatocellular carcinoma; HEM, haemangioma; HAE, hepatic alveolar echinococcosis.
The cut-off value, sensitivity, specificity, positive predictive value and negative predictive value of sST2 on POD1 in predicting PHLF ≥ grade B were 3700 pg/ml, 92%, 85%, 64% and 97% respectively (Table 4).
Table 4.
Cut-off value and performance of sST2 levels in POD1 in PHLF ≥ grade B*
| Items | Whole cohort (173) | Major resection (75) |
|---|---|---|
| PHLF ≥ grade B | n = 39 | n = 33 |
| Cut-off value | 3700 | 3700 |
| Sensitivity (%) | 92.31 | 90.91 |
| Specificity (%) | 85.07 | 78.57 |
| PPV (%) | 64.29 | 76.92 |
POD, postoperative day; PHLF, posthepatectomy liver failure; sST2, soluble suppression of tumourigenicity 2; PPV, predictive positive value; PNV, predictive negative value. *PHLF was graded according to the criteria issued by the International Study Group of Liver Surgery (ISGLS).
Patients with a plasma sST2 level <3700 pg/ml were defined as sST2low, and patients with an sST2 level >3700 pg/ml as sST2high (Table 5 and Table S4). Patients in the sST2high group more frequently suffered from poor prognosis. In the whole and major resection cohorts, 64.3% (n = 36) and 76.9% (n = 30) of sST2high patients respectively, experienced PHLF ≥ grade B while only 2.6% (n = 3) and 8.3% (n = 3) of sST2low patients experienced PHLF ≥ grade B (all P < 0.001) (Table 5). All three grades of PHLF were significantly more frequent in the sST2high group (P < 0.001).
Table 5.
Comparison of sST2high and sST2low groups
| Items | sST2high (n = 56) |
sST2low (n = 117) |
P |
|---|---|---|---|
| Duration of hospital stay (days) median (i.q.r.) | 23.5 (17.0–39.0) | 15.0 (11.0–19.5) | <0.001 |
| Duration of postoperative hospital stay (days) median (i.q.r.) | 12.50 (8.25–22.75) | 7 (6–11) | <0.001 |
| PHLF grade | 0.000 | ||
| ≥B | 36 (64.29) | 3 (2.56) | |
| <B | 20 (35.71) | 114 (97.44) | |
| Clavien–Dindo classification | 0.001 | ||
| Without complication | 34 (60.71) | 95 (81.20) | |
| <IIIa | 9 (16.07) | 16 (13.68) | |
| ≥IIIa | 13 (23.21) | 6 (5.13) |
Values are n (%) unless otherwise stated. sST2, soluble suppression of tumourigenicity 2; i.q.r., interquartile range; PHLF, posthepatectomy liver failure.
Operation time, total and postoperative duration of hospital stay were significantly higher in sST2high than in the sST2low group (all P < 0.001), as shown in Table 5. A total of 25.4% (n = 44) patients experienced complications: sST2high group showed higher postoperative morbidity rate (Clavien-Dindo grades IIIa or higher, Fig. S4e) and all four cases of death (2.3%) were from this group.
Discussion
Despite the development of liver surgery, PHLF still occurs in a significant number of patients who were carefully selected for liver resection. Plasma sST2 levels may represent a good biomarker for predicting PHLF in the immediate postoperative hours. In this study, it was the only independent circulating predictor of PHLF found on POD1; its predictive value remained high regardless of the underlying disease, extent of hepatectomy, presence or absence of liver cirrhosis, HBV infection or steatosis.
ST2 was considered as an ‘orphan’ receptor for many years; IL-33 has only been recognized recently as its only known ligand23. ST2 presents two main isoforms: the membrane-bound form which plays a role in Th2-mediated diseases24 and a soluble form secreted by endothelial and epithelial cells and fibroblasts in response to stress and/or inflammation, used as a heart failure biomarker12,25. Very few studies are available involving liver dysfunction17. This study was the first to look at the status of the IL-33/ST2 axis after hepatectomy. Although both IL-33 and sST2 were elevated, only sST2 was significantly associated with PHLF within the first 24 h posthepatectomy. A wide range of changes in sST2 levels were observed after hepatectomy and a rapid normalization in patients without PHLF. When applying a cut-off value of 3700 pg/ml, sST2 levels on POD1 seemed to have a good clinical performance as evidenced by worse outcomes in the sST2high group.
After tissue injury, IL-33 is released from damaged or necrotic tissue cells and acts as a cellular alarmin for the immune system and is reported to be a proregenerative factor26. In the acetaminophen-induced liver injury (AILI) model, IL-33 deficiency enhanced hepatocyte autophagy and tissue injury27 and from this perspective, it might be a potential biomarker for predicting PHLF. In this study the predictive value of plasma IL-33 in the whole cohort and in the major resection cohort was only 0.587 and 0.557 respectively, which is consistent with previous studies28,29. Although Pan et al. demonstrated the good performance of IL-33 in predicting the prognosis of patients with HCC30, the mRNA level of ST2 but not IL-33 was significantly increased in patients with HCC compared with the healthy control group. Moreover, high heterogeneity of patients and different types of surgery may lead to poor comparability with this study.
With the assistance of the IL-1 receptor accessory protein (IL-1RAcP), ST2L binds with IL-33 and plays a role in different immunoregulatory pathways, which is determined by cell location. sST2, located in the extracellular matrix13,31 and immune cells32, downregulates the IL-33/ST2L axis. After liver resection, the remnant liver encounters an increase in blood flow and vascular endothelial cell injury occurs. Shear stress stimulates the massive secretion of sST2 in the early stage after surgery13, which was confirmed in the present study. Similarly, the correlation between sST2 and aminotransferase levels appeared to be significant. This study showed higher sST2 levels on POD1 in HAE and HCC compared with others; influenced by the immune/inflammatory response and the required type of surgical resection.
This study has some limitations that need to be addressed: as a heterogeneous cohort with less than 30% malignant disease included, which does not allow generalization to a Western population, external validation is required. As sST2 was only assessed in patients with liver resection, the specificity of this marker should be measured in other types of major surgery, such as major abdominal surgery. The long duration of this study and batch effect of the ELISA kit may lead to incorrect results, although the sST2 factor is stable if stored at −20°C for at least 18 months; a standard measurement of plasma sST2 needs to be completed. The deeper regulating mechanisms of this biomarker remain unsolved. Moreover, other possible causes of sST2 elevation in surgical patients were excluded in this study.
A certain level of inflammation is needed to spur effective regeneration, but if there is too much inflammation the liver fails to regenerate. The present study cannot differentiate the inflammatory response to liver regenerative response, therefore, IL-33 and ST2 co-localized with classic inflammatory or regenerative markers should be given full consideration in future studies. Besides, due to a wide range of IL-33/ST2 axis sources, an in-depth study should be focused on the relation between extrahepatic inflammation and liver regeneration to unveil the possible underlying mechanism.
sST2 showed superior predictive performance of PHLF ≥ grade B as early as POD1 for patients undergoing liver resection. sST2 could be a sensitive candidate that may provide fast testing to predict early PHLF.
Supplementary Material
Acknowledgements
J.W. and S.A. contributed equally to this work.
Contributor Information
Jing Wu, Department of Liver Transplantation & Laparoscopic Surgery, Digestive & Vascular Surgery Center, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China.
Shadike Apaer, Department of Liver Transplantation & Laparoscopic Surgery, Digestive & Vascular Surgery Center, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China.
Xiapukaiti Fulati, Department of Liver Transplantation & Laparoscopic Surgery, Digestive & Vascular Surgery Center, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China.
Dominique A Vuitton, WHO Collaborating Centre for Prevention and Treatment of Human Echinococcosis and French National Centre for Echinococcosis, University Bourgogne Franche-Comté and National Reference Centre for Echinococcosis/EurEchino Network, Besançon, France.
Yunfei Zhang, Department of Liver Transplantation & Laparoscopic Surgery, Digestive & Vascular Surgery Center, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China.
Jiangduosi Payiziwula, Department of Liver Transplantation & Laparoscopic Surgery, Digestive & Vascular Surgery Center, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China.
Nuerzhatijiang Anweier, Department of Liver Transplantation & Laparoscopic Surgery, Digestive & Vascular Surgery Center, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China.
Tao Li, Department of Liver Transplantation & Laparoscopic Surgery, Digestive & Vascular Surgery Center, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China.
Kahaer Tuerxun, Department of Hepatobiliary and Pancreatic Surgery, The First People’s Hospital of Kashi Perfecture, Kashi, China.
Tuerganaili Aji, State Key Laboratory of Pathogenesis, Prevention, Treatment of High Incidence Diseases in Central Asia, Xinjiang Medical University, Urumqi, China.
Jinming Zhao, Department of Liver Transplantation & Laparoscopic Surgery, Digestive & Vascular Surgery Center, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China.
Yingmei Shao, State Key Laboratory of Pathogenesis, Prevention, Treatment of High Incidence Diseases in Central Asia, Xinjiang Medical University, Urumqi, China.
Tuerhongjiang Tuxun, Department of Liver Transplantation & Laparoscopic Surgery, Digestive & Vascular Surgery Center, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China; State Key Laboratory of Pathogenesis, Prevention, Treatment of High Incidence Diseases in Central Asia, Xinjiang Medical University, Urumqi, China.
Hao Wen, Department of Liver Transplantation & Laparoscopic Surgery, Digestive & Vascular Surgery Center, The First Affiliated Hospital of Xinjiang Medical University, Urumqi, China; State Key Laboratory of Pathogenesis, Prevention, Treatment of High Incidence Diseases in Central Asia, Xinjiang Medical University, Urumqi, China.
Funding
This study was supported by grants from the National Natural Science Foundation of China (82270632); National Natural Science Foundation of China (82260411); State Key Laboratory of Pathogenesis, Prevention, Treatment of High Incidence Diseases in Central Asia Fund Special Project for Echinococcosis (SKL-HIDCS-2020-BC2); Shu-Lan Excellent Project-Support Program for Overseas Study of Young Talents in Organ Transplantation, 2017.
Disclosure
The authors declare no conflict of interest.
Supplementary material
Supplementary material is available at BJS Open online.
Data availability
Raw data are available from the corresponding authors upon reasonable request.
Author contributions
Jing Wu (Data curation, Formal analysis, Investigation, Software, Validation, Visualization, Writing—original draft), Apaer Shadike (Data curation, Formal analysis, Investigation, Methodology, Software, Validation, Visualization, Writing—original draft), Xiapukaiti Fulati (Data curation, Investigation, Software, Validation, Visualization), Dominique Vuitton (Conceptualization, Methodology, Supervision, Writing—review & editing), Yunfei Zhang (Data curation, Investigation, Validation, Visualization), Jiangduosi Payiziwula (Data curation, Investigation, Software, Validation, Visualization), Nuerzhatijiang Anweier (Data curation, Investigation, Software, Validation, Visualization), Tao Li (Conceptualization, Investigation, Methodology, Resources, Supervision, Writing—review & editing), Kahaer Tuerxun (Conceptualization, Investigation, Methodology, Resources, Software, Supervision), Aji Tuerganaili (Conceptualization, Methodology, Resources, Supervision, Writing—review & editing), Jinming Zhao (Conceptualization, Methodology, Resources, Supervision), Yingmei Shao (Conceptualization, Methodology, Project administration, Resources, Supervision, Writing—review & editing), Tuerhongjiang Tuxun (Conceptualization, Funding acquisition, Methodology, Project administration, Resources, Supervision, Writing—review & editing) and Hao Wen (Conceptualization, Methodology, Project administration, Resources, Supervision, Validation, Writing—review & editing)
References
- 1. Furrer K, Deoliveira ML, Graf R, Clavien PA. Improving outcome in patients undergoing liver surgery. Liver Int 2007;27:26–39 [DOI] [PubMed] [Google Scholar]
- 2. Skrzypczyk C, Truant S, Duhamel A, Langlois C, Boleslawski E, Koriche D et al. Relevance of the ISGLS definition of posthepatectomy liver failure in early prediction of poor outcome after liver resection: study on 680 hepatectomies. Ann Surg 2014;260:865–870; discussion 870 [DOI] [PubMed] [Google Scholar]
- 3. Schmidt LE, Larsen FS. MELD score as a predictor of liver failure and death in patients with acetaminophen-induced liver injury. Hepatology 2007;45:789–796 [DOI] [PubMed] [Google Scholar]
- 4. Pagliaro L. MELD: the end of Child-Pugh classification? J Hepatol 2002;36:141–142 [DOI] [PubMed] [Google Scholar]
- 5. DuBray BJ Jr, Levy RV, Balachandran P, Conzen KD, Upadhya GA, Anderson CD et al. Novel three-dimensional imaging technique improves the accuracy of hepatic volumetric assessment. HPB (Oxford) 2011;13:670–674 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Lesurtel M, Belghiti J. Temporary portal vein embolization as a starter of liver regeneration. J Hepatol 2008;49:313–315 [DOI] [PubMed] [Google Scholar]
- 7. Aji T, Dong JH, Shao YM, Zhao JM, Li T, Tuxun T et al. Ex vivo liver resection and autotransplantation as alternative to allotransplantation for end-stage hepatic alveolar echinococcosis. J Hepatol 2018;69:1037–1046 [DOI] [PubMed] [Google Scholar]
- 8. McPhail MJ, Kriese S, Heneghan MA. Current management of acute liver failure. Curr Opin Gastroenterol 2015;31:209–214 [DOI] [PubMed] [Google Scholar]
- 9. Ringe B, Xiao G, Sass DA, Karam J, Shang S, Maroney TP et al. Rescue of a living donor with liver transplantation. Am J Transplant 2008;8:1557–1561 [DOI] [PubMed] [Google Scholar]
- 10. Cayrol C, Girard JP. Interleukin-33 (IL-33): a nuclear cytokine from the IL-1 family. Immunol Rev 2018;281:154–168 [DOI] [PubMed] [Google Scholar]
- 11. Mehraj V, Ponte R, Routy JP. The dynamic role of the IL-33/ST2 axis in chronic viral-infections: alarming and adjuvanting the immune response. EBioMedicine 2016;9:37–44 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Mueller T, Jaffe AS. Soluble ST2–analytical considerations. Am J Cardiol 2015;115:8B–21B [DOI] [PubMed] [Google Scholar]
- 13. Wang Y, Tan X, Gao H, Yuan H, Hu R, Jia L et al. Magnitude of soluble ST2 as a novel biomarker for acute aortic dissection. Circulation 2018;137:259–269 [DOI] [PubMed] [Google Scholar]
- 14. Sugano A, Seo Y, Ishizu T, Sai S, Yamamoto M, Hamada-Harimura Y et al. Soluble ST2 and brain natriuretic peptide predict different mode of death in patients with heart failure and preserved ejection fraction. J Cardiol 2019;73:326–332 [DOI] [PubMed] [Google Scholar]
- 15. Wang M, Shen G, Xu L, Liu X, Brown JM, Feng D et al. IL-1 receptor like 1 protects against alcoholic liver injury by limiting NF-κB activation in hepatic macrophages. J Hepatol 2018;68:109–117 [DOI] [PubMed] [Google Scholar]
- 16. Yin H, Huang BJ, Yang H, Huang YF, Xiong P, Zheng F et al. Pretreatment with soluble ST2 reduces warm hepatic ischemia/reperfusion injury. Biochem Biophys Res Commun 2006;351:940–946 [DOI] [PubMed] [Google Scholar]
- 17. Roth GA, Zimmermann M, Lubsczyk BA, Pilz J, Faybik P, Hetz H et al. Up-regulation of interleukin 33 and soluble ST2 serum levels in liver failure. J Surg Res 2010;163:e79–e83 [DOI] [PubMed] [Google Scholar]
- 18. Bossuyt PM, Reitsma JB, Bruns DE, Gatsonis CA, Glasziou PP, Irwig L et al. STARD 2015: an updated list of essential items for reporting diagnostic accuracy studies. BMJ 2015;351:h5527. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Hamaguchi M, Kojima T, Itoh Y, Harano Y, Fujii K, Nakajima T et al. The severity of ultrasonographic findings in nonalcoholic fatty liver disease reflects the metabolic syndrome and visceral fat accumulation. Am J Gastroenterol 2007;102:2708–2715 [DOI] [PubMed] [Google Scholar]
- 20. Johnson PJ, Berhane S, Kagebayashi C, Satomura S, Teng M, Reeves HL et al. Assessment of liver function in patients with hepatocellular carcinoma: a new evidence-based approach-the ALBI grade. J Clin Oncol 2015;33:550–558 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Strasberg SM. Nomenclature of hepatic anatomy and resections: a review of the Brisbane 2000 system. J Hepatobiliary Pancreat Surg 2005;12:351–355 [DOI] [PubMed] [Google Scholar]
- 22. Rahbari NN, Garden OJ, Padbury R, Brooke-Smith M, Crawford M, Adam R et al. Posthepatectomy liver failure: a definition and grading by the International Study Group of Liver Surgery (ISGLS). Surgery 2011;149:713–724 [DOI] [PubMed] [Google Scholar]
- 23. Kakkar R, Lee RT. The IL-33/ST2 pathway: therapeutic target and novel biomarker. Nat Rev Drug Discov 2008;7:827–840 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Bloodworth MH, Rusznak M, Bastarache L, Wang J, Denny JC, Peebles RS Jr. Association of ST2 polymorphisms with atopy, asthma, and leukemia. J Allergy Clin Immunol 2018;142:991–993.e3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Weinberg EO, Shimpo M, Hurwitz S, Tominaga S, Rouleau JL, Lee RT. Identification of serum soluble ST2 receptor as a novel heart failure biomarker. Circulation 2003;107:721–726 [DOI] [PubMed] [Google Scholar]
- 26. Wen Y, Emontzpohl C, Xu L, Atkins CL, Jeong JM, Yang Y et al. Interleukin-33 facilitates liver regeneration through serotonin-involved gut-liver axis. Hepatology 2023;77:1580–1592 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Wang Z, Wu L, Pan B, Chen Y, Zhang T, Tang N. Interleukin 33 mediates hepatocyte autophagy and innate immune response in the early phase of acetaminophen-induced acute liver injury. Toxicology 2021;456:152788. [DOI] [PubMed] [Google Scholar]
- 28. Bergis D, Kassis V, Ranglack A, Koeberle V, Piiper A, Kronenberger B et al. High serum levels of the interleukin-33 receptor soluble ST2 as a negative prognostic factor in hepatocellular carcinoma. Transl Oncol 2013;6:311–318 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Sun Z, Chang B, Huang A, Hao S, Gao M, Sun Y et al. Plasma levels of soluble ST2, but not IL-33, correlate with the severity of alcoholic liver disease. J Cell Mol Med 2019;23:887–897 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Pan X, Liu J, Li M, Liang Y, Liu Z, Lao M et al. The association of serum IL-33/ST2 expression with hepatocellular carcinoma. BMC Cancer 2023;23:704. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Miller AM, Xu D, Asquith DL, Denby L, Li Y, Sattar N et al. IL-33 reduces the development of atherosclerosis. J Exp Med 2008;205:339–346 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Sciatti E, Merlo A, Scangiuzzi C, Limonta R, Gori M, D'Elia E et al. Prognostic value of sST2 in heart failure. J Clin Med 2023;12:3970. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Raw data are available from the corresponding authors upon reasonable request.