Abstract
Background/Aim
Extreme hyperbilirubinemia is occasionally observed in intensive care unit (ICU) and non-ICU settings. This study examined the etiologies of extreme hyperbilirubinemia (bilirubin level ≥12 mg/dL) and the factors associated with the 30-day mortality.
Methods
This retrospective observational cohort study identified 439 patients with extreme hyperbilirubinemia at the Gyeongsang National University Changwon Hospital between 2016 and 2020. The patients were classified into three groups and 11 diseases according to their etiology. The risk factors associated with 30-day mortality at the baseline were investigated using the Cox proportional hazards model.
Results
Of 439 patients with extreme hyperbilirubinemia, 287, 78, and 74 were in the liver cirrhosis/malignancy group, the ischemic injury group, and the benign hepatobiliary-pancreatic etiological group, respectively, with corresponding 30-day mortality rates of 42.9%, 76.9%, and 17.6%. The most common disease leading to hyperbilirubinemia was a pancreatobiliary malignancy (28.7%), followed by liver cirrhosis (17.3%), hepatocellular carcinoma (10.9%), and liver metastases (8.4%). The etiologies of hyperbilirubinemia, obstructive jaundice, infection, albumin level, creatinine level, and prothrombin time-international normalized ratio were independently associated with the 30-day mortality.
Conclusions
This study suggests three etiologies of extreme hyperbilirubinemia in the ICU and non-ICU settings. The prognosis of patients with extreme hyperbilirubinemia depends largely on the etiology and the presence of obstructive jaundice.
Keywords: Hyperbilirubinemia, Etiology, Obstructive jaundice, Ischemic hepatitis, Malignancy
INTRODUCTION
Various intra- and extrahepatic factors can cause new-onset jaundice in adults. A retrospective study of adult inpatients and outpatients with new-onset jaundice showed that 55% had an intrahepatic etiology (including viral hepatitis, alcoholic liver disease, autoimmune hepatitis [AIH], drug-induced liver injury [DILI], and decompensation of pre-existing chronic liver disease). The remaining 45% had an extrahepatic etiology (including gallstone disease, sepsis, hemolysis, and extrahepatic malignancy).1
The overall mortality rate of patients with jaundice varied according to the etiology of the hyperbilirubinemia.1-4 Sepsis/ shock is associated with a high mortality rate (33–63%) and usually occurs in intensive care unit (ICU) settings.1,2,4 The decompensation of pre-existing chronic liver disease and hepatobiliary-pancreatic (HBP) malignancy has a high mortality rate of 17–42%.1-3 In contrast, hepatocellular injury (such as viral hepatitis, DILI, and alcoholic hepatitis) and benign pancreatobiliary (PB) disease (such as gallstone disease and pancreatitis) have relatively favorable prognoses; most of these patients can be managed in non-ICU settings.1,2,4
The sequential organ failure assessment (SOFA) scoring system, which is used to diagnose organ failure in ICU, classifies hyperbilirubinemia from 0 (normal) to 4 (total bilirubin level ≥12 mg/dL).5 The chronic liver failure-SOFA score, a modified SOFA score for patients with chronic liver disease, defines liver failure as a serum total bilirubin level of ≥12 mg/dL.6 Extreme hyperbilirubinemia (total bilirubin level ≥12 mg/dL) is commonly encountered in ICU settings and is often observed in patients with sepsis, advanced cirrhosis, and HBP malignancy.4,7 Han et al.4 conducted a retrospective study of 427 patients with extreme hyperbilirubinemia who were admitted to the ICU. On the other hand, extreme hyperbilirubinemia is occasionally observed in patients with benign HBP diseases such as viral hepatitis, AIH, DILI, and gallstone disease.2,3 Clinicians often encounter patients with extreme hyperbilirubinemia in ICU and non-ICU settings. This study evaluated the etiologies of extreme hyperbilirubinemia in ICU and non-ICU patients and investigated the factors related to the 30-day mortality.
SUBJECTS AND METHODS
1. Study population
A retrospective electronic medical chart review was performed on 472 adult patients (>18 years) with serum bilirubin levels ≥12 mg/dL admitted to a secondary referral hospital from January 2016 to December 2020. The exclusion criteria were a failure to classify the etiology because they were lost to follow-up (n=13) and a lack of available imaging, clinical, and laboratory data (n=20). The remaining 439 patients with extreme hyperbilirubinemia were included in the study (Fig. 1). The Institutional Review Board of Gyeongsang National University Changwon Hospital reviewed and approved this study (IRB File No. 2021-06-032), and this study was conducted in accordance with the 1964 Declaration of Helsinki. The need for informed consent was waived owing to the retrospective design of the study, as determined by the institutional review board.
Fig. 1.
Flow chart of patients with extreme hyperbilirubinemia divided into three etiological groups. HBP, hepatobiliary pancreas; PB, pancreatobiliary; DILI, drug-induced liver injury; AIH, autoimmune hepatitis; LC, liver cirrhosis; HCC, hepatocellular carcinoma.
2. Data collection
After enrollment, clinical, laboratory, and imaging investigations of all patients were conducted via a retrospective chart review to determine the etiology of hyperbilirubinemia. The baseline demographic data were collected, including age, sex, and underlying diseases, such as diabetes, viral hepatitis, liver cirrhosis (LC), HBP malignancy, congestive heart failure, and end-stage renal disease. Clinical data, such as medication and alcohol history, renal replacement therapy (RRT) for acute kidney injury, extracorporeal membrane oxygenation (ECMO), and hepatic decompensation, and laboratory data, such as aspartate aminotransferase (AST), alanine aminotransferase (ALT), total bilirubin, albumin, alkaline phosphatase (ALP), prothrombin time-international normalized ratio (PT-INR), and creatinine, were extracted.
3. Definitions and classification of hyperbilirubinemia
Extreme hyperbilirubinemia was defined as at least one episode of a serum bilirubin level of ≥12 mg/dL by retrospective chart review. If the patients experienced multiple episodes of extreme hyperbilirubinemia, only the first episode was considered for statistical convenience. The index date for analyzing the baseline clinical and laboratory data was defined as the first date of serum bilirubin levels ≥12 mg/dL. Obstructive jaundice is defined as an impediment to bile flow into the duodenum because of an anatomical obstruction of the extrahepatic biliary passage. Hepatic decompensation was defined as the presence of ascites, hepatorenal syndrome, hepatic encephalopathy, or documented gastroesophageal variceal hemorrhage.
Two experienced hepatologists classified patients with extreme hyperbilirubinemia into three groups comprising 11 diseases according to etiology (Fig. 1): (1) benign HBP group, including PB disease, DILI/AIH, viral hepatitis, and alcoholic hepatitis; (2) LC/malignancy group, including LC, PB malignancy, hepatocellular carcinoma (HCC), and liver metastasis; and (3) ischemic injury group, including postoperative jaundice, ischemic hepatitis, and sepsis-induced cholestasis. Each disease is defined in upplementary Table 1. If a patient had two or more etiologies, the single etiology with a more significant effect on hyperbilirubinemia was selected based on the judgment of two experienced hepatologists. Hemolysis was excluded as an etiology of extreme hyperbilirubinemia because it infrequently causes a bilirubin level of >5 mg/dL unless accompanied by renal disease, liver disease, or severe hematologic malignancy. Recovery from extreme hyperbilirubinemia was defined as a decrease in bilirubin level <3 mg/dL within 30 days.
4. Statistical analysis
All data handling and statistical analyses were performed using PASW Statistics software version 18 (SPSS Inc., Chicago, IL, USA). The baseline characteristics of the patients are expressed as numbers (%) for categorical data and median (interquartile range) for continuous data. The intergroup differences in categorical data were measured using a Mann– Whitney U test, whereas a Kruskal–Wallis test was used for continuous data. The survival curves for 30-day mortality were constructed using the Kaplan–Meier method with a log-rank test. The risk factors associated with the 30-day mortality at the baseline were determined using the Cox proportional hazards model. For statistical convenience, this study excluded recovery from hyperbilirubinemia during the course of jaundice within 30 days of evaluating the risk factors at the baseline. Univariate and multivariate analyses were performed to evaluate a hazard ratio (HR) and 95% confidence interval (CI) for 30-day mortality. All statistical tests were performed using two-tailed p-value tests with a statistical significance level of 0.05.
RESULTS
1. Patient characteristics
The baseline characteristics of the 439 patients with extreme hyperbilirubinemia were classified into three main etiological groups (Table 1): LC/malignancy (n=287), ischemic injury (n=78), and benign HBP (n=74). The median age of the patients was 64 years, and 341 were male (71.5%). In the cohort of patients, 122, 68, and seven had LC, diabetes, and end-stage renal disease, respectively. The median baseline bilirubin level was 15 mg/dL. Of the 439 patients, 96 (21.9%) were admitted to the ICU, and 136 (31.0%), 41 (9.3%), and 34 (7.8%) developed hepatic decompensation, underwent RRT at the baseline, and underwent ECMO, respectively.
Table 1.
Baseline characteristics of patients with extreme hyperbilirubinemia
| Characteristics | Benign HBP group | LC/malignancy group | Ischemic injury group | Total | p-value |
|---|---|---|---|---|---|
| No. | 74 (16.9) | 287 (65.4) | 78 (17.8) | 439 | |
| Age, year | 57.0 (47.8–69.3) | 65.0 (54.0–75.0) | 67.5 (55.0–77.0) | 64.0 (52.0–75.0) | 0.009 |
| Male gender | 50 (67.6) | 207 (72.1) | 57 (73.1) | 341 (71.5) | 0.700 |
| BMI (kg/m2) | 23.7 (21.6–26.7) | 22.3 (19.5–25.2) | 24.0 (21.1–29.7) | 22.9 (20.0–26.2) | <0.001 |
| LCa | 7 (9.5) | 106 (36.9) | 9 (11.5) | 122 (27.8) | <0.001 |
| Diabetes | 16 (21.6) | 42 (14.6) | 10 (12.8) | 68 (15.5) | 0.258 |
| ESRD | 0 | 3 (1.0) | 4 (5.1) | 7 (1.6) | 0.230 |
| ECMO | 0 | 1 (0.3) | 33 (42.3) | 34 (7.8) | <0.001 |
| Hepatic decompensation | 12 (16.2) | 117 (40.8) | 7 (9.0) | 136 (31.0) | <0.001 |
| Infection | 16 (21.6) | 92 (32.1) | 51 (65.4) | 159 (36.2) | <0.001 |
| Admission at ICU | 9 (12.2) | 18 (6.3) | 69 (88.5) | 96 (21.9) | <0.001 |
| Obstructive jaundice | 19 (25.7) | 121 (42.2) | 0 | 140 (31.9) | <0.001 |
| RRT | 3 (4.1) | 10 (3.5) | 28 (35.9) | 41 (9.3) | <0.001 |
| Bilirubin, mg/dL | 14.9 (13.0–19.7) | 15.9 (13.7–21.3) | 13.6 (12.6–14.6) | 15.0 (13.2–19.5) | <0.001 |
| AST, U/L | 224.5 (115.8–648.0) | 150.0 (93.0–275.0) | 173.0 (80.8–612.3) | 157.0 (93.0–327.0) | 0.004 |
| ALT, U/L | 199.5 (73.5–496.3) | 73.0 (37.0–148.0) | 87.5 (46.8–331.3) | 87.0 (41.0–203.0) | <0.001 |
| Albumin, g/dL | 3.3 (2.8–3.8) | 2.8 (2.4–3.3) | 2.6 (2.4–2.8) | 2.8 (2.5–3.4) | <0.001 |
| ALP, U/L | 195.0 (133.0–324.5) | 323.0 (177.0–587.0) | 161.5 (112.8–256.8) | 251.0 (146.0–479.0) | <0.001 |
| LDH, U/L | 303.0 (242.5–428.0) | 283.0 (213.0–457.0) | 824.0 (341.1–2262.0) | 299.5 (222.8–515.5) | <0.001 |
| Creatinine, mg/dL | 0.88 (0.66–1.09) | 0.87 (0.68–1.35) | 1.31 (0.93–1.88) | 0.94 (0.70–1.39) | <0.001 |
| PT-INR | 1.17 (0.98–1.43) | 1.31 (1.07–1.93) | 1.45 (1.26–1.77) | 1.31 (1.09–1.78) | <0.001 |
Data are presented as the number (%) for categorical data and medians (interquartile range) for continuous data.
HBP, hepatobiliary-pancreatic; LC, liver cirrhosis; BMI, body mass index; ESRD, end-stage renal disease; ECMO, extracorporeal membrane oxygenation; ICU, intensive care unit; RRT, renal replacement therapy; AST, aspartate aminotransferase; ALT, alanine aminotransferase; ALP, alkaline phosphatase; LDH, lactate dehydrogenase; PT-INR, prothrombin time-international normalized ratio.
p-value: The Kruskal–Wallis test and Mann–Whitney U test.
aThe mean Child-Pugh score in patients with LC was 9.9, 11.0, and 10.2 in benign HBP group, LC/malignancy group, and ischemic hepatitis group, respectively.
2. Etiological groups
Fig. 1 shows the three etiological groups and 11 subdivided diseases according to the etiology of patients with extreme hyperbilirubinemia. A PB malignancy was the most common etiology of extreme hyperbilirubinemia, accounting for 28.7% of all cases, followed by LC (17.3%), HCC (10.9%), and liver metastasis (8.4%). The patients in the LC/malignancy group had significantly higher baseline bilirubin and ALP levels than those in the benign HBP and ischemic injury groups (Table 1). On the other hand, AST, ALT, and albumin levels were highest in the benign HBP group. The ischemic injury group included patients with ECMO (42.3%), concomitant infection (65.4%), ICU admission (88.5%), and RRT (35.9%); these rates were higher than those in the LC/malignancy and benign HBP groups. Several terminally ill cancer patients who declined ICU treatment were included in the LC/malignancy group.
3. Clinical outcome
The 30-day mortality rate was 44.6% for the entire study group. Fig. 2 shows the impact of etiology on mortality. The ischemic injury, LC/malignancy, and benign HBP groups had 30-day mortality rates of 76.9%, 42.9%, and 17.6%, respectively (Fig. 3A). Interestingly, the 30-day mortality rate of the patients was closely related to the etiology of the extreme hyperbilirubinemia. For example, patients with ischemic hepatitis, postoperative jaundice, and sepsis-induced cholestasis in the ischemic injury group had high 30-day mortality rates of 84.0%, 78.3%, and 70.0%, respectively. In contrast, patients with alcoholic hepatitis, DILI/AIH, viral hepatitis, and PB disease in the benign HBP group had relatively favorable 30-day mortality rates of 22.2%, 21.7%, 15.4%, and 10.0%, respectively (Table 2). In the LC/malignancy group, patients with HCC (79.2%) and liver metastasis (75.7%) had a higher 30-day mortality rate than those with LC (34.2%) and PB malignancies (24.6%).
Fig. 2.
Thirty-day mortality according to the etiology. HBP, hepatobiliary pancreas; PB, pancreatobiliary; DILI, drug-induced liver injury; AIH, autoimmune hepatitis; LC, liver cirrhosis; HCC, hepatocellular carcinoma.
Fig. 3.
Kaplan–Meier curves for the 30-day survival according to (A) etiological group and (B) presence of obstructive jaundice. HBP, hepatobiliary pancreas, LC, liver cirrhosis.
Table 2.
Clinical outcome of patients with extreme hyperbilirubinemia
| Etiological group | Disease | N (%) | 30-day mortality | Obstructive jaundice | Recovery from hyperbilirubinemia |
|---|---|---|---|---|---|
| Benign HBP | PB disease | 20 (4.6) | 2 (10.0) | 19 (95.0) | 16 (80.0) |
| DILI/AIH | 23 (5.2) | 5 (21.7) | 0 | 13 (56.5) | |
| Viral hepatitis | 13 (3.0) | 2 (15.4) | 0 | 7 (53.8) | |
| Alcoholic hepatitis | 18 (4.1) | 4 (22.2) | 0 | 9 (50.0) | |
| LC/malignancy | LC | 76 (17.3) | 26 (34.2) | 0 | 11 (14.5) |
| PB malignancy | 126 (28.7) | 31 (24.6) | 109 (86.5) | 55 (43.7) | |
| HCC | 48 (10.9) | 38 (79.2) | 5 (10.4) | 1 (2.1) | |
| Liver metastasis | 37 (8.4) | 28 (75.7) | 7 (18.9) | 5 (13.5) | |
| Ischemic injury | Postoperative jaundice | 23 (5.2) | 18 (78.3) | 0 | 2 (8.7) |
| Ischemic hepatitis | 25 (11.8) | 21 (84.0) | 0 | 3 (12.0) | |
| Sepsis-induced cholestasis | 30 (6.8) | 21 (70.0) | 0 | 7 (23.3) |
Values are presented as number (%).
HBP, hepatobiliary-pancreatic; PB, pancreatobiliary; DILI, drug-induced liver injury; AIH, autoimmune hepatitis; LC, liver cirrhosis; HCC, hepatocellular carcinoma.
One hundred and forty patients with obstructive jaundice were confirmed. Obstructive jaundice was observed in 109 patients (109/126, 86.5%) with a PB malignancy, 19 patients with PB disease (19/20, 95%), seven patients (7/37, 18.9%) with liver metastasis, and five patients (5/48, 10.4%) with HCC. Patients with obstructive jaundice (15.0%) had a significantly lower mortality rate than those without obstructive jaundice (58.5%, p<0.001) (Fig. 3B).
Recovery (bilirubin level <3 mg/dL) from extreme hyperbilirubinemia within 30 days occurred in 129 (29.4%) patients: 72 (25.1%), 45 (60.8%), and 12 (15.8%) patients in the LC/malignancy, benign HBP, and ischemic injury groups, respectively. The recovery rate from extreme hyperbilirubinemia was highest in patients with PB disease (80%), followed by those with DILI/AIH (56.5%), viral hepatitis (53.8%), alcoholic hepatitis (50%), and PB malignancy (43.7%) (Table 2).
4. Impact of the presence of obstruction and recovery from hyperbilirubinemia
Patients with obstructive jaundice (72/140, 51.4%) had significantly higher recovery from hyperbilirubinemia than those without obstructive jaundice (56/299, 18.7%). Patients who recovered from hyperbilirubinemia within 30 days (2/129, 1.6%) had a significantly lower 30-day mortality rate than those who did not recover (194/310, 62.6%). Hence, whether the bilirubin level recovers to less than 3 mg/dL during jaundice therapy has a significant impact on the mortality of patients with extreme hyperbilirubinemia. In addition, the most effective recovery from hyperbilirubinemia was observed in patients with obstructive jaundice who underwent biliary decompression with successful interventions (e.g., endoscopic retrograde cholangiography and percutaneous transhepatic biliary drainage). For example, although the LC/malignancy group had a poor prognosis, patients (19/121, 15.7%) with obstructive jaundice treated with effective decompression showed a significantly lower 30-day mortality rate than those with non-obstructive jaundice (104/166, 62.7%).
5. Predictors associated with 30-day mortality
The benign HBP group was used as a reference for the etiological groups in the Cox proportional hazards model to assess the independent risk factors associated with 30-day mortality. Non-obstructive jaundice (adjusted HR=2.42, 95% CI=1.38–4.24), infection (adjusted HR=1.41, 95% CI=1.02– 1.96), albumin level (adjusted HR=0.48, 95% CI=0.35–0.66), creatinine level (adjusted HR=1.13, 95% CI=1.05–1.21), PT-INR (adjusted HR=1.16, 95% CI=1.03–1.30), LC/malignancy group (adjusted HR=2.30, 95% CI=1.22–4.34), and ischemic injury group (adjusted HR=4.02, 95% CI=2.05–7.88) were independently associated with the 30-day mortality in multivariate analysis (Table 3).
Table 3.
Risk factors associated with the 30-day mortality in univariate and multivariate Cox regression analysis
| Variable | Univariate analysis | Multivariate analysis | ||
|---|---|---|---|---|
| p-value | HR (95% CI) | p-value | HR (95% CI) | |
| Male | 0.026 | 1.46 (1.05–2.04) | 0.224 | 1.27 (0.87–1.85) |
| BMI (kg/m2) | 0.009 | 1.04 (1.01–1.07) | 0.254 | 1.02 (0.99–1.05) |
| Etiological group | ||||
| Benign HBP group | Reference | Reference | ||
| LC/malignancy group | <0.001 | 2.87 (1.62–5.09) | 0.010 | 2.30 (1.22–4.34) |
| Ischemic injury group | <0.001 | 7.76 (4.25–14.16) | <0.001 | 4.02 (2.05–7.88) |
| Non-obstructive jaundice | <0.001 | 5.39 (3.42–8.48) | 0.002 | 2.42 (1.38–4.24) |
| Infection | <0.001 | 2.09 (1.58–2.77) | 0.039 | 1.41 (1.02–1.96) |
| Hepatic decompensation | <0.001 | 1.88 (1.41–2.49) | 0.990 | 1.00 (0.68–1.46) |
| Albumin, g/dL | <0.001 | 0.31 (0.25–0.40) | <0.001 | 0.48 (0.35–0.66) |
| Creatinine, mg/dL | <0.001 | 1.24 (1.17–1.31) | 0.001 | 1.13 (1.05–1.21) |
| PT-INR | <0.001 | 1.31 (1.23–1.40) | 0.011 | 1.16 (1.03–1.30) |
HR, hazard ratio; CI, confidence interval; BMI, body mass index; HBP, hepatobiliary pancreas; LC, liver cirrhosis; PT-INR, prothrombin time-international normalized ratio.
DISCUSSION
This study examined the etiology and prognosis of extreme hyperbilirubinemia (bilirubin level of ≥12 mg/dL), which is used to define liver failure based on chronic liver failure-SOFA.6 Extreme hyperbilirubinemia was classified into three groups comprising 11 diseases according to etiology. The most frequent disease with extreme hyperbilirubinemia was PB malignancy (28.7%), followed by LC (17.3%), HCC (10.9%), and liver metastasis (8.4%), which belonged to the LC/malignancy group. The 30-day mortality rate for the entire patient cohort was 44.6%: 76.9%, 42.9%, and 17.6% for the ischemic injury, LC/malignancy, and benign HBP groups, respectively.
Patients who developed hyperbilirubinemia due to benign hepatocellular, pancreatic, or biliary diseases, regardless of underlying LC, were classified into the benign HBP group. All patients with obstructive jaundice in the benign HBP group had PB diseases, such as cholecystitis, cholangitis, and pancreatitis. Among patients with PB disease, 80% recovered from hyperbilirubinemia, and the 30-day mortality rate was only 10%. Patients with obstructive jaundice had a relatively favorable prognosis because hyperbilirubinemia can be improved by biliary decompression. The remaining patients without PB diseases in the benign HBP group, who had conditions, such as DILI/AIH, viral hepatitis, or alcoholic hepatitis, did not exhibit biliary obstruction. Half of these patients recovered from hyperbilirubinemia within 30 days, and the mortality rate was approximately 20%. In a previous study from the authors’ center, the 30-day mortality rates of patients with benign pancreatobiliary and hepatocellular diseases with aminotransferase levels >400 U/L were 1.7% and 3.2%, respectively.8 Other studies reported a mortality rate of 0–7% in patients with benign pancreatobiliary and hepatocellular disease.9-11 Hence, patients with benign HBP disease accompanied by extreme hyperbilirubinemia have a poorer prognosis than those without extreme hyperbilirubinemia.
Patients with pre-existing chronic liver disease that affects liver function (e.g., cirrhosis and intrahepatic and extrahepatic malignancies) were classified into the LC/malignancy group. In the LC malignancy group, 86% of patients with a PB malignancy had a biliary obstruction, and 43% of patients recovered from hyperbilirubinemia through decompression. Nevertheless, most patients with HCC and liver metastasis had non-obstructive jaundice caused by liver failure associated with tumor infiltration. Patients with HCC and liver metastasis had a poor prognosis, with only 7% recovering from hyperbilirubinemia and a 78% 30-day mortality rate. In addition, all patients with LC had non-obstructive jaundice due to liver failure associated with hepatic decompensation. Only 14% of patients with LC recovered from hyperbilirubinemia, and 34% of patients with LC died within 30 days. Therefore, most patients with pre-existing chronic liver disease accompanied by extreme hyperbilirubinemia caused by liver failure or decompensation (without biliary obstruction) die within 30 days. In this study, patients with pre-existing LC who developed benign PB disease were classified into the benign HBP group based on the leading cause of hyperbilirubinemia.
None of the patients in the ischemic injury group had a biliary obstruction. Among the 23 patients with postoperative jaundice, 12 developed hyperbilirubinemia after cardiovascular surgery, and six developed hyperbilirubinemia after gastrointestinal surgery. The conditions contributing to extreme hyperbilirubinemia in most patients with postoperative jaundice were related to ischemic injury caused by sepsis, hypotension, and hypoxemia. Sepsis-induced cholestasis was defined as hyperbilirubinemia due to impaired bile flow during or following the septic process without a sharp increase in aminotransferase levels and not meeting the criteria for ischemic hepatitis. Patients in the ischemic injury group, including those with postoperative jaundice, ischemic hepatitis, and sepsis-induced cholestasis, had a poorer prognosis, with a 30-day mortality rate of up to 77%. Multiple organ failure and underlying diseases contributed heavily to the prognosis of patients with ischemic hepatitis. Previous studies reported a 40–50% mortality rate of patients with ischemic hepatitis, regardless of hyperbilirubinemia.12-14 In other studies, the mortality rate in patients with postoperative jaundice (bilirubin level >3 mg/dL) was 10–20%. These findings suggest that patients with ischemic injury accompanied by extreme hyperbilirubinemia have poorer prognosis than those without extreme hyperbilirubinemia.
Recently, a Korean study reported the prognosis of critically ill patients with extreme hyperbilirubinemia (bilirubin level of ≥12 mg/dL) who were admitted to the ICU.4 They reported a 76.1% mortality rate, which is higher than the 44.1% mortality rate observed in this study. In contrast to the present study, they reported a correlation between the peak bilirubin levels and mortality. The reason for these differences is that this study included patients admitted to non-ICU settings, including those with benign HBP disease and malignancy, who recovered from hyperbilirubinemia through decompression. In addition, this study included critically ill patients with malignancies who refused ICU admission.
This study had some limitations. This study was retrospective in data collection. Therefore, there may have been limitations in determining the etiology of hyperbilirubinemia. The study design of determining the etiology was based on the judgment of two hepatologists, which might have imposed a misclassification bias on the cohort. In addition, this was a single secondary referral hospital experience, which might not reflect the general etiology of hyperbilirubinemia nationwide. Nevertheless, this study evaluated the characteristics and prognoses of patients with diverse etiologies of extreme hyperbilirubinemia, including non-ICU patients.
In conclusion, extreme hyperbilirubinemia is occasionally observed in ICU and non-ICU settings and may be associated with high mortality due to various etiologies. These observations suggest that it is important to establish the etiology in patients with extreme hyperbilirubinemia and confirm the presence of biliary obstruction.
SUPPLEMENTARY MATERIAL
Supplementary material is available at the Korean Journal of Gastroenterology website (https://www.kjg.or.kr/).
Footnotes
Financial support
None.
Conflict of interest
None.
REFERENCES
- 1.Vuppalanchi R, Liangpunsakul S, Chalasani N. Etiology of new-onset jaundice: how often is it caused by idiosyncratic drug-induced liver injury in the United States? Am J Gastroenterol. 2007;102:558–562. quiz 693. doi: 10.1111/j.1572-0241.2006.01019.x. [DOI] [PubMed] [Google Scholar]
- 2.Whitehead MW, Hainsworth I, Kingham JG. The causes of obvious jaundice in South West Wales: perceptions versus reality. Gut. 2001;48:409–413. doi: 10.1136/gut.48.3.409. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Björnsson E, Ismael S, Nejdet S, Kilander A. Severe jaundice in Sweden in the new millennium: causes, investigations, treatment and prognosis. Scand J Gastroenterol. 2003;38:86–94. doi: 10.1080/00365520310000492. [DOI] [PubMed] [Google Scholar]
- 4.Han HS, Park CM, Lee DS, Sinn DH, Gil E. Evaluating mortality and recovery of extreme hyperbilirubinemia in critically ill patients by phasing the peak bilirubin level: A retrospective cohort study. PLoS One. 2021;16:e0255230. doi: 10.1371/journal.pone.0255230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Vincent JL, Moreno R, Takala J, et al. The SOFA (Sepsis-related Organ Failure Assessment) score to describe organ dysfunction/failure. On behalf of the Working Group on Sepsis-Related Problems of the European Society of Intensive Care Medicine. Intensive Care Med. 1996;22:707–710. doi: 10.1007/BF01709751. [DOI] [PubMed] [Google Scholar]
- 6.Moreau R, Jalan R, Gines P, et al. Acute-on-chronic liver failure is a distinct syndrome that develops in patients with acute decompensation of cirrhosis. Gastroenterology. 2013;144:1426–1437.:1437.e1-9. doi: 10.1053/j.gastro.2013.02.042. [DOI] [PubMed] [Google Scholar]
- 7.Kwo PY, Cohen SM, Lim JK. ACG Clinical Guideline: Evaluation of abnormal liver chemistries. Am J Gastroenterol. 2017;112:18–35. doi: 10.1038/ajg.2016.517. [DOI] [PubMed] [Google Scholar]
- 8.Kwak JY, Kim HG, Han JH, Jeon H, Cha RR, Lee SS. Association of the etiology and peak level of markedly elevated aminotransferases with mortality: a multicenter study. Hepatol Commun. 2023;7:e0149. doi: 10.1097/HC9.0000000000000149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Whitehead MW, Hawkes ND, Hainsworth I, Kingham JG. A prospective study of the causes of notably raised aspartate aminotransferase of liver origin. Gut. 1999;45:129–133. doi: 10.1136/gut.45.1.129. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Galvin Z, McDonough A, Ryan J, Stewart S. Blood alanine aminotransferase levels >1,000 IU/l - causes and outcomes. Clin Med (Lond) 2015;15:244–247. doi: 10.7861/clinmedicine.15-3-244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Björnsson HK, Olafsson S, Bergmann OM, Björnsson ES. A prospective study on the causes of notably raised alanine aminotransferase (ALT) Scand J Gastroenterol. 2016;51:594–600. doi: 10.3109/00365521.2015.1121516. [DOI] [PubMed] [Google Scholar]
- 12.Van den Broecke A, Van Coile L, Decruyenaere A, et al. Epidemiology, causes, evolution and outcome in a single-center cohort of 1116 critically ill patients with hypoxic hepatitis. Ann Intensive Care. 2018;8:15. doi: 10.1186/s13613-018-0356-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Aboelsoud MM, Javaid AI, Al-Qadi MO, Lewis JH. Hypoxic hepatitis- its biochemical profile, causes and risk factors of mortality in critically-ill patients: A cohort study of 565 patients. J Crit Care. 2017;41:9–15. doi: 10.1016/j.jcrc.2017.04.040. [DOI] [PubMed] [Google Scholar]
- 14.Tapper EB, Sengupta N, Bonder A. The incidence and outcomes of ischemic hepatitis: A systematic review with meta-analysis. Am J Med. 2015;128:1314–1321. doi: 10.1016/j.amjmed.2015.07.033. [DOI] [PubMed] [Google Scholar]