Abstract
Background
Potential indications for surgery frequently arise in patients awaiting liver transplantation. There is a risk of hepatic decompensation and death triggered by surgical trauma, but this has not been studied in detail in this unique population. We aimed to quantify the impact of surgical interventions in patients awaiting liver transplantation on hepatic function and identify risk factors for decompensation.
Methods
All surgeries between 2000 and 2018 in patients awaiting liver transplantation in a highvolume German liver transplant center were analyzed retrospectively. Change in liver function measured as indicated by MELD score was assessed and complication rates recorded. The primary endpoint was a composite of an increase in MELD score by > 5 points or death. A logistic regression model was used for multivariate analysis to identify risk factors.
Results
In total, 177 surgical procedures in 148 patients were analyzed. The primary endpoint was reached in 42 cases (23.7%). The overall in-hospital complication rate (including death) was 44.1%. Multivariate analysis identified elevated leukocyte count, perioperative blood transfusion, preoperative presence of ascites, and preoperative circulatory support as independent risk factors for a decline in liver function or death.
Conclusion
Surgery in patients awaiting liver transplantation carries a relevant risk of hepatic decompensation and death that needs to be considered when deciding whether to perform elective surgery prior to or defer until after liver transplantation.
Keywords: cirrhosis, liver transplantation, surgery, waiting list, MELD
Abbreviations: ACLF, Acute-on-Chronic Liver Failure; ALT, Alanine transaminase; aPTT, Activated partial thromboplastin time; ASA, American Society of Anesthesiologists physical status classification system; AST, Aspartate transaminase; CHE, Cholinesterase; CRP, C-reactive Protein; dl, Deciliter; i.e., id est; g, Gram; Hb, Hemoglobin; HCC, Hepatocellular carcinoma; INR, International Normalized Ratio; log, Logarithm; l, Liter; MELD, Model for End-stage Liver Disease; mg, Milligram; min, Minutes; mmol, Millimole; OR, Odds Ratio; SBP, Spontaneous Bacterial Peritonitis; Tsd, Thousand; vs, Versus; μl, Microliter
During the evaluation process for liver transplantation or while on the waiting list, indications for surgical interventions may arise. However, perioperative complications may be more common in individuals with advanced liver disease, and surgical trauma may trigger hepatic decompensation. In some instances, the indication for surgery may be absolute and mandate immediate surgical intervention. In all other instances, the medical team caring for the patient must decide whether to proceed with surgery or postpone until after the liver transplantation.
Liver cirrhosis has been known to be a risk factor for perioperative adverse and serious adverse events.1, 2, 3, 4, 5, 6, 7, 8 The morbidity and mortality risks correlate with the severity of the cirrhosis measured by CHILD or MELD score.2, 3, 4, 5,7, 8, 9, 10, 11, 12, 13, 14 Cirrhotic patients undergoing surgery are more likely to develop bleeding due to coagulopathy, hepatic decompensation, or even acute-on-chronic liver failure (ACLF).2,3,13,15
So far, several studies have investigated the perioperative mortality of patients with liver cirrhosis, but only little is known specifically about the risk for patients on the waiting list for liver transplantation and about the impact of surgeries on liver function.6,8, 9, 10,16, 17, 18, 19 The study of Nyberg et al. demonstrated a higher rate of hepatic decompensation in cirrhotic patients (12.8%) undergoing orthopedic surgery than in cirrhotic controls (4.9%) who did not have surgery.15 Although del Olmo et al. showed that the significantly higher rate of complications in cirrhotic patients compared to noncirrhotic controls after surgery is mainly due to cirrhosis-associated complications, no information on the liver function development due to these complications has been provided.7
Patients listed for liver transplantation are a unique patient cohort compared to cirrhotic patients in general. Since patients are only listed for liver transplantation if improvement of life expectancy or quality of life is expected, they show a unique set of characteristics. Patients are usually younger than the average cirrhotic patient, have fewer comorbidities, and are under close medical supervision at their local transplant center. The last point is also the reason why indications for surgery might arise more likely than in other cirrhotic patients. The underlying liver disease is either more severe or deteriorates faster than in other cirrhotic patients. Additionally, not all patients waiting for liver transplantation suffer from liver cirrhosis. Polycystic liver disease and malignant tumors are examples of reasons for patients requiring liver transplantation without suffering from cirrhosis. To our knowledge, the existing studies on mortality on the waiting list for liver transplantation do not analyze the impact of surgical procedures, and a study on the mortality of cirrhotic patients undergoing surgery with patients waiting for liver transplantation among them does not analyze these patients as a subgroup.17,20, 21, 22 There is one study by Mozzati et al. covering super pulsed laser therapy after dental extraction in 12 patients waiting for liver transplantation with the focus on the efficiency of the new therapy approach and not on the impact of the procedure itself on the patients and their liver function.23
The importance of preventing deterioration of liver function in patients awaiting liver transplantation is presented in a study by Merion et al., showing a ΔMELD greater than 5 (an increase of 5 MELD points or more) over a 30-day period is associated with a three times greater mortality risk for patients on the waiting list for liver transplantation.24
In most studies analyzing cirrhotic patients undergoing surgery, only certain major or elective surgeries are selected.3,5,8, 9, 10, 11, 12,14,16,19 Minor procedures like port system implantations, abdominal wall hernia repair, or cholecystectomies are not discussed, although they frequently occur and include potential perioperative and postoperative risks as well.
In this project, we investigated the impact of surgeries on the liver function of patients on the waiting list for liver transplantation. Furthermore, we analyzed complication rates and identified potential risk factors for reaching the primary endpoint, which was defined as a composite of a ΔMELD >5 or death, in order to provide more information on the safety of surgery in this group of patients.
Methods
Patients
For this retrospective study, all patients listed on the waiting list for liver transplantation at Hanover Medical School, Germany, between January 2000 and August 2018 were identified. Inclusion criteria for our analysis were as follows: age older than eighteen years, surgery at our hospital while being on the waiting list or during the evaluation progress before being listed, and consent to the use of pseudonymized clinical data for research purposes. Exclusion criteria were prior liver transplantation, high urgency listing for liver transplantation, and missing laboratory values for MELD score calculation.
Eurotransplant provided information on names, date of birth, sex, time on the waiting list, and reason for the end of the listing. The clinical records were examined for surgical procedures during the time on the waiting list. If the patients had surgery, data on demographics, surgical procedure, preoperative laboratory values, and postoperative period were collected from the patient’s clinical chart.
Surgery
All interventions with a surgical protocol were considered as operations. For patients with several surgeries, surgeries concerning the same indication as previously performed surgeries already included were not recorded separately.
The urgency status of the surgeries, i.e. emergent or elective, was drawn from the surgical records. All surgeries in the abdominal or thoracic cavity, thyroid surgery, surgery on fractured bones, vascular surgery on arteries, surgery on large joints, and neurosurgical procedures were classified as major. Minor surgery included drainage or catheter surgeries, surgery in the oral or nasal cavity, abdominal or thoracic wall surgery, minor trauma, ophthalmological, gynecologic and urologic procedures, tracheostomy creation, skin and wound surgeries, surgery on hemorrhoids, and vascular surgery for hemodialysis.
The use of erythrocyte concentrates was noted as transfusion.
Mortality after surgery was defined as the death of the patient within the first 14 days after surgery. In-hospital complications were noted and stratified by the Clavien-Dindo Classification.25 For every case, only the most severe complication was noted.
Liver Function and Acute-on-Chronic Liver Failure
For evaluating the development of the liver function, the MELD score was calculated at two time points before (“Baseline MELD” and “Pre-OP MELD”) and two time points after surgery (“Post-OP MELD” and “Late MELD”).
-
•
“Baseline MELD” was calculated two months before the surgical procedure (range 6 months to 14 days before surgery).
-
•
“Pre-OP MELD” was calculated based on the latest available values prior to the start of the surgical procedure (range 14 days before surgery to just before surgery).
-
•
“Post-OP MELD” was calculated based on the earliest available values 24 h after the end of the surgical procedure (range 24 h to 14 days after the surgery).
-
•
“Late MELD” was calculated two months after the surgical procedure (range 14 days to 6 months after the surgery).
ΔMELD was calculated between “Late MELD” and “Baseline MELD” in order to observe the long-term effect of surgery on the liver function. Data for calculating “Baseline MELD” and “Late MELD” were collected as close as possible to two months before or after surgery, resulting in an ideal time frame of four months for ΔMELD. The primary endpoint was defined as a composite endpoint of ΔMELD >5 or death. In order to also portray patients who died in the first 14 days after surgery in Figure 1, Figure 3, “Post-OP MELD” and “Late MELD” were set to 40 for these patients. This was only done if the reason for missing “Post-OP MELD” or “Late MELD” was death in the first 14 days and only for creating Figure 1, Figure 3.
Figure 1.
A: MELD development before and after surgery. “Baseline MELD.” 2 months (14 days–6 months) before surgery; “Pre-OP MELD”: right before surgery (up to 14 days before); “Post-OP MELD”: seven days (24 h–14 days) after surgery; “Late MELD”: two months (14 days–6 months) after surgery. B: ΔMELD for all patients and deaths in the perioperative period. ΔMELD: Difference between “Late” and “Baseline MELD.”
Figure 3.
ΔMELD is shown as mean and standard deviation stratified by different factors. A: “Baseline MELD”; B: Extent of surgery; C: Urgency of surgery; D: Pre-existing Ascites; E: Perioperative Transfusion; F: Leukocytes.
ACLF was assessed using the grading system by Moreau et al.26
Statistics
All statistical analyses were performed using SPSS software (version 25.0, Chicago, IL, USA). Two-sided Fisher’s exact test was used to analyze categorical data and Mann–Whitney test for continuous data. Multivariate analysis was performed using a logistic regression model in order to identify independent risk factors for worsening of liver function.
Ethics
The ethics committee of Hanover Medical School approved this study on August 14, 2018 (Nr. 8026_BO_K_2018).
All authors had access to the data and approved the final manuscript.
Results
From January 2000 to August 2018, 177 surgeries were performed in 148 patients. Baseline characteristics of all 177 cases are shown in Table 1. The median age was 50, ranging from 21 to 60, and in 48% of the cases, the patient was female. 10.15% of patients were diabetics accounting for 18.1% (n = 32) of surgeries; 21.62% of patients were active smokers accounting for 22.6% (n = 40) of surgeries. Reasons for listing for liver transplantation were: autoimmune liver disease (28.2%), viral hepatitis (23.7%), alcoholic liver disease (18.6%), metabolic diseases (10.2%), and cryptogenic liver cirrhosis (7.3%). In 27.7% (n = 49) the “Baseline MELD” score was 10 or lower, 26.6% of the cases (n = 47) had a “Baseline MELD” between 11 and 15, 24.3% (n = 43) between 16 and 20 and in 21.5% (n = 38) of the cases the “Baseline MELD” score was 21 or higher. The median “Baseline MELD” was 15.
Table 1.
Patient Baseline Characteristics.
| Median | Range | |
|---|---|---|
| Age | 50 | 21–66 |
| n | % of total | |
| Sex | ||
| Women | 85 | 48 |
| Men | 92 | 52 |
| Etiology | ||
| Autoimmune | 50 | 28.2 |
| Viral | 42 | 23.7 |
| Alcoholic | 33 | 18.6 |
| Metabolic | 18 | 10.2 |
| Other | 16 | 9.0 |
| Cryptogenic | 13 | 7.3 |
| Multiple | 5 | 2.8 |
| Baseline MELD | ||
| ≤10 | 49 | 27.7 |
| 11–15 | 47 | 26.6 |
| 16–20 | 43 | 24.3 |
| ≥21 | 38 | 21.5 |
| Diabetes | ||
| Yes | 32 | 18.1 |
| No | 145 | 81.9 |
| Nicotine use | ||
| Never smoked | 66 | 37.3 |
| Former smoker | 21 | 11.9 |
| Smoker | 40 | 22.6 |
| No Data | 50 | 28.2 |
MELD Model for end-stage liver disease.
Major surgery was performed in 45.2% (n = 80) and minor surgery in 54.8% (n = 98). In 24.3% (n = 43) the procedure was emergent while 75.7% (n = 134) of the surgeries had an elective status. Surgery was performed under general anesthesia in 71.2% (n = 126), local or regional anesthesia was used in 27.7% (n = 49) and in 1.1% (n = 2) of the procedures conscious sedation was used. The surgeries took 60 min or less in 46.9% (n = 83), 60–120 min in 31.6% (n = 56) and more than 120 min in 20.3% (n = 36). Of the surgeries 42.4% (n = 75) were abdominal, 9.6% (n = 17) cardiothoracic, 6.8% (n = 12) trauma/orthopedic, 18.6% (n = 33) otolaryngologic/dental and 22.6% (n = 40) belonged to other fields (Table 2).
Table 2.
Procedures Analyzed.
| n | % of total | |
|---|---|---|
| Extent of surgery | ||
| Major | 80 | 45.2 |
| Minor | 97 | 54.8 |
| Type of anesthesia | ||
| General | 126 | 71.2 |
| Local or regional | 49 | 27.7 |
| Conscious sedation | 2 | 1.1 |
| Duration of surgery | ||
| ≤60 min | 83 | 46.9 |
| 61–120 min | 56 | 31.6 |
| >120 min | 36 | 20.3 |
| No Data | 2 | 1.1 |
| Urgency of surgical procedure | ||
| Elective | 134 | 75.7 |
| Emergent | 43 | 24.3 |
| Type of surgery | ||
| Abdominal | 75 | 42.4 |
| Otolaryngology/Dentistry | 33 | 18.6 |
| Cardiothoracic | 17 | 9.6 |
| Trauma/Orthopedic | 12 | 6.8 |
| Other | 40 | 22.6 |
In Figure 1A, the development of the MELD score before and after surgery is shown for all cases with available data. Patients, who died within the first 14 days after surgery, thus lacking values for “Late MELD” or “Post-OP MELD,” are represented with a MELD score of 40. The mean “Baseline MELD” was 16 ± 6.85, raising to a mean “Pre MELD” of 18 ± 8.6. After surgery, the mean “Post-OP MELD” was 19 ± 9.1, and the mean “Late MELD” score 19 ± 10.4. Figure 1B gives an overview of the ΔMELD between the “Late MELD” and “Baseline MELD” score for each case, including the cases in which the patient died in the postoperative period. In 31.1% (n = 55) of the cases the MELD score improved, and for 11.9% (n = 21) of the patients, the MELD score did not change. The MELD score deteriorated in 47.5% (n = 84) and 9.6% (n = 17) of the patients died in the first 14 postoperative days. The most frequent causes of death were hepatic failure, sepsis, and death due to cardiovascular reason.
In-hospital complications, including the death of a patient, occurred in 44.1% of all surgeries, as shown in Figure 2. The most frequent complications were wound complications, infection/sepsis, hepatic decompensation, and other complications. For 15.8% of the cases, complications without the need for surgical, endoscopic, or radiological interventions occurred (Clavien-Dindo Grades 1 and 2). In 13.6% of the surgeries, surgical, endoscopic, or radiological interventions were needed (Clavien-Dindo Grades 3a and 3b), and 3.4% of life-threatening complications required intensive care (Clavien-Dindo-Grades 4a and 4b). Overall, in 11.3% of all cases, the patient died in hospital after surgery. Minor surgeries had a higher rate without complications than major surgeries (70.1% vs. 38.8%, p < 0.001). Only 1% of minor surgeries led to complications requiring intensive care, while 6.3% of major surgeries led to these complications (p = 0.092). Patients died more often after major surgeries than after minor surgeries (18.8% vs. 5.2%, p = 0.007)). Elective surgeries were more frequently performed without complications than emergent surgeries (63.4% vs. 32.6%, p = 0.001). Intensive care because of complications was necessary after only 0.7% of all elective surgeries and after 11.6% of emergent surgeries (p = 0.003). In-hospital death was less frequent in elective surgeries compared to emergent surgeries (6.7% vs. 25.6%, p = 0.002).
Figure 2.
In-hospital complication rate for all procedures, extent, and urgency of the surgery.
Figure 3 gives an overview of the ΔMELD’s for every case, stratified by “Baseline MELD” (3A), the extent of surgery (3B), the urgency of surgery (3C), pre-existing ascites (3D), need for perioperative transfusions (3E) and leukocyte count (3F). For patients who died within the first 14 days after surgery, thus lacking values for “Late MELD,” calculations were performed with a “Late MELD” of 40. There was no significant difference between the mean ΔMELD’s of patients with a “Baseline MELD” lower or higher than 15 and of surgeries classified as minor or major. Emergent surgeries had a significantly higher mean ΔMELD than elective surgeries. Mean ΔMELD was also significantly higher in patients with pre-existing ascites, need for perioperative transfusions, and in patients with a leukocyte count higher than 10,000/μl.
Overall the primary endpoint (ΔMELD >5 or perioperative death) was reached after 23.7% (n = 42) of surgical procedures analyzed. Univariate analysis of potential risk factors for reaching the primary endpoint is shown in Table 3. Smoking, hemodialysis, pre-existing ascites or hepatic encephalopathy, preoperative mechanical ventilation, and circulatory support were significant preoperative factors for worsening of liver function or death in the univariate analysis. With regard to surgical factors, patients requiring transfusions, emergent surgeries, or surgeries shorter than 90 min were significantly more likely to have a ΔMELD higher than 5 or to die. While the “Baseline MELD” had no significant correlation with the development of the liver function, a “Pre-OP MELD” score of 15 or higher was a risk factor for worsening of liver function or death in the univariate analysis. Laboratory values were categorized and showed that high bilirubin level, high INR, high leukocyte and low thrombocyte count, low hemoglobin level, prolonged aPTT, high CRP, and low CHE level correlated with a ΔMELD over 5 or death of the patient. The univariate analysis also identified an ACLF grade of 2 or 3 as a risk factor for worsening of liver function or death of patients compared to no ACLF or ACLF grade 1.
Table 3.
Univariate and Multivariate Analysis.
| Univariate Analysis |
Multivariate Analysis |
|||||
|---|---|---|---|---|---|---|
| Risk Factor | Stable liver functiona (total n = 135) n (% of total) | Perioperative deteriorationb or death (total n = 42) n (% of total) | p-value | OR | 95%-CI | p-value |
| Gender | ||||||
| Male | 69 (51.1%) | 23 (54.8%) | 0.73 | |||
| Female | 66 (48.9%) | 19 (48.9%) | ||||
| Age | ||||||
| ≤50 | 69 (51.1%) | 25 (59.5%) | 0.38 | |||
| >50 | 66 (48.9%) | 17 (40.5%) | ||||
| Diabetes | ||||||
| Yes | 26 (19.3%) | 6 (14.3%) | 0.65 | |||
| No | 109 (80.7%) | 36 (85.7%) | ||||
| Smoking | ||||||
| Not smoking | 44 (32.6%) | 22 (52.4%) | 0.02 | |||
| Smoking/Former smoker | 52 (38.5%) | 9 (21.4%) | ||||
| Hemodialysis | ||||||
| Yes | 11 (8.2%) | 10 (23.8%) | 0.01 | |||
| No | 121 (89.6%) | 31 (73.8%) | ||||
| Esophageal varices | ||||||
| Yes | 75 (55.5%) | 27 (64.3%) | 0.37 | |||
| No | 60 (45.5%) | 15 (35.7%) | ||||
| History of Variceal Hemorrhage | ||||||
| Yes | 35 (25.9%) | 14 (33.3%) | 0.43 | |||
| No | 100 (74.1%) | 28 (66.7%) | ||||
| History of SBP | ||||||
| Yes | 27 (20%) | 12 (28.6%) | 0.29 | |||
| No | 108 (80%) | 30 (71.4%) | ||||
| HCC | ||||||
| Yes | 20 (14.8%) | 4 (9.5%) | 0.45 | |||
| No | 115 (85.2%) | 38 (90.5%) | ||||
| Extent | ||||||
| Minor | 77 (57%) | 20 (47.6%) | 0.29 | |||
| Major | 58 (43%) | 22 (52.4%) | ||||
| Urgency | ||||||
| Elective | 113 (83.7%) | 21 (50%) | <0.001 | 1.1 | 0.4–3.4 | 0.805 |
| Emergent | 22 (16.3%) | 21 (50%) | ||||
| Duration of surgery | ||||||
| ≤90 min | 81 (60%) | 32 (76.2%) | 0.04 | |||
| >90 min | 53 (39.3%) | 9 (21.4%) | ||||
| Anesthesia | ||||||
| General | 91 (67.4%) | 35 (83.3%) | 0.05 | |||
| Local/Regional/Conscious sedation | 44 (32.6%) | 7 (16.7%) | ||||
| ASA | ||||||
| 2–3 | 52 (38.5%) | 13 (31%) | 0.4 | |||
| 4 | 6 (4.4%) | 3 (7.1%) | ||||
| Pre-existing Ascites | ||||||
| Yes | 34 (25.2%) | 25 (59.5%) | <0.001 | 5.2 | 2.1–12.9 | <0.001 |
| No | 101 (74.8%) | 17 (40.5%) | ||||
| Pre-existing Hepatic Encephalopathy | ||||||
| Yes | 15 (11.1%) | 12 (28.6%) | 0.01 | |||
| No | 120 (88.9%) | 30 (71.4%) | ||||
| Mechanical Ventilation | ||||||
| Yes | 2 (1.5%) | 8 (19%) | <0.001 | |||
| No | 133 (98.5%) | 34 (81%) | ||||
| Circulatory Support | ||||||
| Yes | 1 (0.7%) | 10 (23.8%) | <0.001 | 12.4 | 1.3–118.4 | 0.029 |
| No | 134 (99.3%) | 32 (76.2%) | ||||
| Transfusion perioperative | ||||||
| Yes | 12 (8.9%) | 16 (38.1%) | <0.001 | 3.8 | 1.2–12.4 | 0.025 |
| No | 123 (91.1%) | 26 (61.9%) | ||||
| Baseline MELD | ||||||
| ≤15 | 74 (54.8%) | 22 (52.4%) | 0.86 | |||
| >15 | 61 (45.2%) | 20 (47.6%) | ||||
| Pre-OP MELD | ||||||
| <15 | 55 (40.7%) | 7 (16.7%) | 0.001 | |||
| ≥15 | 57 (42.2%) | 31 (73.8%) | ||||
| Sodium | ||||||
| ≤135 mmol/l | 36 (26.7%) | 16 (38.1%) | 0.06 | |||
| >135 mmol/l | 86 (63.7%) | 17 (40.5%) | ||||
| Bilirubin | ||||||
| ≤1.2 mg/dl | 39 (28.9%) | 4 (9.5%) | 0.004 | |||
| >1.2 mg/dl | 74 (54.8%) | 35 (83.3%) | ||||
| Creatinin | ||||||
| ≤1.2 mg/dl | 88 (65.2%) | 21 (50%) | 0.08 | |||
| >1.2 mg/dl | 39 (28.9%) | 18 (42.9%) | ||||
| INR | ||||||
| ≤1.2 | 45 (33.3%) | 3 (7.1%) | 0.001 | |||
| >1.2 | 84 (62.2%) | 36 (85.7%) | ||||
| Leukocytes | ||||||
| ≤10 Tsd/μl | 110 (81.5%) | 20 (47.6%) | <0.001 | 4.1 | 1.5–11.2 | 0.006 |
| >10 Tsd/μl | 18 (13.3%) | 20 (47.6%) | ||||
| Hb | ||||||
| <10 g/dl | 46 (34.1%) | 27 (64.3%) | 0.001 | |||
| ≥10 g/dl | 82 (60.7%) | 13 (31%) | ||||
| Thrombocytes | ||||||
| <100 Tsd/μl | 57 (42.2%) | 28 (66.7%) | 0.006 | |||
| ≥100 Tsd/μl | 71 (52.6%) | 12 (28.6%) | ||||
| aPTT | ||||||
| <40 s | 72 (53.3%) | 13 (31%) | 0.02 | |||
| ≥40 s | 54 (40%) | 24 (57.1%) | ||||
| CRP | ||||||
| ≤5 mg/l | 44 (32.6%) | 1 (2.4%) | <0.001 | |||
| >5 mg/l | 76 (56.3%) | 36 (85.7%) | ||||
| AST | ||||||
| <35 U/l | 35 (25.9%) | 7 (16.7%) | 0.28 | |||
| ≥35 U/l | 78 (57.8%) | 28 (66.7%) | ||||
| ALT | ||||||
| <45 U/l | 77 (57%) | 27 (64.3%) | 0.84 | |||
| ≥45 U/l | 35 (25.9%) | 11 (26.2%) | ||||
| Alkaline Phosphatase | ||||||
| ≤130 U/l | 36 (26.7%) | 10 (23.8%) | >0.99 | |||
| >130 U/l | 44 (32.6%) | 13 (31%) | ||||
| Gamma Glutamyltransferase | ||||||
| ≤55 U/l | 30 (22.2%) | 10 (23.8%) | >0.99 | |||
| >55 U/l | 67 (49.6%) | 22 (52.4%) | ||||
| CHE | ||||||
| ≤4.26 kU/l | 46 (34.1%) | 16 (38.1%) | 0.047 | |||
| >4.26 kU/l | 28 (20.7%) | 2 (4.8%) | ||||
| Pre ACLF | ||||||
| 0–1 | 116 (85.9%) | 27 (64.3%) | <0.001 | |||
| 2–3 | 5 (3.7%) | 11 (26.2%) | ||||
ACLF Acute-on-Chronic Liver Failure, ALT Alanine transaminase, aPTT Activated partial thromboplastin time, ASA American Society of Anesthesiologists physical status classification system, AST: Aspartate transaminase, CHE Cholinesterase, CI Confidence Interval, CRP C-reactive Protein, Hb Hemoglobin, HCC Hepatocellular carcinoma, INR International Normalized Ratio, MELD Model for End-stage Liver Disease, OR Odds Ratio, SBP Spontaneous Bacterial Peritonitis.
ΔMELD<5.
ΔMELD>5.
Regarding the rather small group of deaths and cases with ΔMELD higher than 5, only five significant factors of the univariate analysis were included in the multivariate analysis shown in Table 3. We chose factors (perioperative transfusion, preoperative leukocyte count, preoperative ascites, emergent vs. elective surgery, and preoperative circulatory support) that were considered to be plausible determinants of liver function development by the investigators based on the literature (Neef et al., An et al., del Olmo et al., Lee et al., Neef et al., Meunier et al., Telem et al.) and their clinical judgment.1,3,7,16,18,27,28 The logistic regression identified perioperative transfusion, high preoperative leukocyte count, pre-existing ascites, and preoperative circulatory support as independent risk factors for worsening of liver function or death, while the urgency of surgery did not show a significant impact.
Discussion
With the exception of highly emergent procedures, surgery can often be scheduled before or postponed until after liver transplantation. The risk of postponing surgery needs to be balanced against the risk of surgical trauma triggering a potentially critical decline in liver function or even ACLF or death. In order to investigate the safety of surgery in patients awaiting liver transplantation, we retrospectively analyzed the change of liver function after surgery in order to identify risk factors associated with perioperative liver function decline.
The median “Baseline MELD” was 15 that matches three-month mortality of 6% (MELD 10–19) according to the findings of Wiesner et al.29
Overall, a relevant decline in liver function defined as ΔMELD > 5 or death, i.e. the primary endpoint of our analysis, occurred in only 42 (23.7%) of the procedures analyzed. In 31.1% of the cases, the MELD score improved after surgery, possibly because the condition leading to the surgery improved afterward and had a beneficial impact on the liver function. Moreover, this favorable outcome may be due in part to the high level of care for patients awaiting liver transplantation and their relatively young age. Nonetheless, factors able to predict who is likely to benefit and who will likely decline are needed in order to decide whether elective surgical procedures in patients awaiting liver transplantation should be performed pretransplant or posttransplant.
Several factors have been identified in our analysis and are discussed in the following paragraphs:
Leukocyte count: an elevated preoperative leukocyte count was an independent risk factor for a ΔMELD greater than 5 or perioperative death. This is consistent with the findings of Telem et al., who showed that the leukocyte count of patients with uncomplicated abdominal surgery was significantly lower than that of patients with complications and of the patients who died or required liver transplantation.28 Conversely, an elevated preoperative leukocyte count was not significantly associated with perioperative mortality after nonhepatic general surgery in the study by Neeff et al.1 Moreau et al. found the leukocyte count in patients with ACLF to be significantly higher than in the patients without ACLF; moreover, the likelihood of death increased with higher leukocyte count.26 Furthermore, Zaccherini et al. describe high leukocyte count to be an independent risk factors for developing nosocomial ACLF and associated with higher mortality in these patients.30 The authors of the “PREDICT” study discovered that very high leukocyte count and serum CRP are associated with developing ACLF and propose inflammation to be one of the causes of ACLF.31 Thus, our finding that elevated leukocyte count is associated with perioperative deterioration of liver function is consistent with the majority of comparable studies.
Blood transfusion: perioperative transfusion was independently associated with reaching the primary endpoint, i.e. worsening of liver function or death. This is in line with other studies that identified blood transfusions as a risk factor for morbidity and mortality in cirrhotic patients. Neeff et al. examined the mortality after 138 general nonhepatic surgeries in cirrhotic patients and showed the need for intraoperative blood transfusion to be an independent risk factor associated with mortality.1 The number of transfused blood units was also an independent risk factor for cirrhosis related complications after nonhepatic surgery in cirrhotic patients in the study by del Olmo et al.7 Lee et al. investigated risk factors for morbidity after colorectal cancer surgery in 161 cirrhotic patients, and intraoperative transfusion was an independent risk factor for postoperative morbidity.16 Thus, the largely consistent observation that blood transfusions are associated with perioperative risk in cirrhotics appears to hold true in the population of patients awaiting liver transplantation.
Ascites: multivariate analysis identified the presence of ascites as an independent risk factor for deterioration of liver function or death. In a study assessing the morbidity and mortality after colorectal surgery in cirrhotic patients, preoperative ascites was a risk factor for postoperative liver-associated complications such as ascites infection, gastrointestinal bleeding, and encephalopathy.27 The presence of ascites was also an independent risk factor for adverse outcome after abdominal surgical procedures in 100 cirrhotic patients in a study by Telem et al.28 A study investigating the impact of ascites on the waitlist mortality found that the mortality was significantly higher in patients with at least moderate ascites (15.4%) compared to patients with mild (6.6%, p < 0.0001) or without ascites (4.1%, p < 0.0001).32 Finally, the “PREDICT” study by Trebicka et al. demonstrated that ascites is associated with systemic inflammation, which, together with changes in portal hypertension, suggested to be the causes of acute decompensation of liver cirrhosis.31
Circulatory failure: not surprisingly, preoperative need for circulatory support was a strong independent risk factor for worsening of liver function or death after surgery (OR 12.4; p = 0.029).
In contrast to other studies in cirrhotic patients, the urgency of the procedure was not an independent risk factor for worsening of liver function.1,3,28 The reason for this difference is not immediately clear, but it may suggest that cirrhotic patients awaiting liver transplantations may differ in important aspects from other cirrhotic populations. Possibly, the lower prevalence of comorbidities and the higher intensity of care offset the negative effects of emergent procedures seen in other populations.
Interestingly, in univariate analysis, surgery shorter than 90 min was a risk factor for deterioration of liver function or death. Possibly, longer surgery is only performed in patients deemed as stable enough and therefore has overall a better outcome. Since we were not able to include this risk factor in multivariate analysis, this would be interesting for further studies.
In terms of surgical complications, the in-hospital complication rate of 44.1% seen in our cohort is comparable to the ones reported by del Olmo et al. (nonhepatic surgery in patients with liver cirrhosis: 50.4%), Lee et al. (colorectal surgery in patients with liver cirrhosis: 37.3%), Telem et al. (abdominal surgery in patients with liver cirrhosis: 43%) and Jang et al. (curative surgery for gastric cancer in patients with liver cirrhosis: 39%).7,16,28,33 A higher complication rate was reported by Meunier et al. with a complication rate of 77% in cirrhotic patients undergoing colorectal surgery.27
Our study has several limitations, most notably its retrospective design. Potentially there could be a selection bias of the patients who had surgery. Moreover, patient data was not always completely documented, so that we had to exclude patients with incomplete data for “Baseline MELD” and “Late MELD” or information on potential risk factors. As there was a wide variety of surgeries with only small patient numbers for each procedure, we were not able to analyze the impact of the type of surgery on the liver function. This could be addressed in future studies, including more patients, for example, as a multicenter study. A prospective study could also examine the prognostic characteristics of other more specific diagnostic tools i.e. LiMAx measurement in patients having surgery while waiting for liver transplantation.34 It would also be interesting to analyze if surgery and possible complications after surgery have an impact on the likelihood of later successful transplantation.
In conclusion, among cirrhotic patients awaiting liver transplantation that undergo surgery, only a minority suffer a perioperative decline in liver function. Independent risk factors for worsening of the liver function were elevated leukocytes, presence of ascites, need for circulatory support, and perioperative blood transfusions. These findings may be helpful in scheduling elective surgical procedures in liver transplant candidates.
CREDIT AUTHORSHIP CONTRIBUTION STATEMENT
I.H. carried out the study and analysis, supported by L.S., F.V., and T.v.H. T.V. guided and supported the statistical analysis. I.H. wrote the manuscript, supported by L.S. and T.v.H. T.v.H and F.V. developed the idea and supervised the project. N.R. and M.M. supervised the study. All authors critically reviewed the manuscript.
Conflicts of interest
The authors have none to declare.
Funding
None.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.jceh.2021.03.011.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
References
- 1.Neeff H., Mariaskin D., Spangenberg H.C., Hopt U.T., Makowiec F. Perioperative mortality after non-hepatic general surgery in patients with liver cirrhosis: an analysis of 138 operations in the 2000s using child and MELD scores. J Gastrointest Surg. 2011;15:1–11. doi: 10.1007/s11605-010-1366-9. [DOI] [PubMed] [Google Scholar]
- 2.Friedmann L.S. The risk of surgery in patients with liver disease. Hepatology. 1999;29:1617–1623. doi: 10.1002/hep.510290639. [DOI] [PubMed] [Google Scholar]
- 3.An Y., Xiao Y., Zhong Q. Open-Heart surgery in patients with liver cirrhosis: indications, risk factors, and clinical outcomes. Eur Surg Res. 2007;39:67–74. doi: 10.1159/000099145. [DOI] [PubMed] [Google Scholar]
- 4.Pandey C.K., Karna S.T., Pandey V.K., Tandon M., Singhal A., Mangla V. Perioperative risk factors in patients with liver disease undergoing non-hepatic surgery. World J Gastrointest Surg. 2012;4:267–274. doi: 10.4240/wjgs.v4.i12.267. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Käser S.A., Hofmann I., Willi N., Stickel F., Maurer C.A. Liver cirrhosis/severe fibrosis is a risk factor for anastomotic leakage after colorectal surgery. Gastroenterol Res Pract. 2016;2016 doi: 10.1155/2016/1563037. Article ID 1563037. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lin C.H., Hsu R.B. Cardiac surgery in patients with liver cirrhosis: risk factors for predicting mortality. World J Gastroenterol. 2014;20:12608–12614. doi: 10.3748/wjg.v20.i35.12608. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.del Olmo J.A., Flor-Lorente B., Flor-Civara B., et al. Risk factors for nonhepatic surgery in patients with cirrhosis. World J Surg. 2003;27:647–652. doi: 10.1007/s00268-003-6794-1. [DOI] [PubMed] [Google Scholar]
- 8.Valmasoni M., Pierobon E.S., De Pasqual C.A., et al. Esophageal cancer surgery for patients with concomitant liver cirrhosis: a single-center matched cohort study. Ann Surg Oncol. 2017;24:763–769. doi: 10.1245/s10434-016-5610-8. [DOI] [PubMed] [Google Scholar]
- 9.Csikesz N.G., Nguyen L.N., Tseng J.F., Shah S.A. Nationwide volume and mortality after elective surgery in cirrhotic patients. J Am Coll Surg. 2009;209:96–103. doi: 10.1016/j.jamcollsurg.2008.09.006. [DOI] [PubMed] [Google Scholar]
- 10.Teh S.H., Nagorney D.M., Stevens S.R., et al. Risk factors for mortality after surgery in patients with cirrhosis. Gastroenterology. 2007;132:1261–1269. doi: 10.1053/j.gastro.2007.01.040. [DOI] [PubMed] [Google Scholar]
- 11.Cohen S.M., Te H.S., Levitsky J. Operative risk of total hip and knee arthroplasty in cirrhotic patients. J Arthroplasty. 2005;20:460–466. doi: 10.1016/j.arth.2004.05.004. [DOI] [PubMed] [Google Scholar]
- 12.Thielmann M., Mechmet A., Neuhäuser M., et al. Risk prediction and outcomes in patients with liver cirrhosis undergoing open-heart surgery. Eur J Cardio Thorac Surg. 2010;38:592–599. doi: 10.1016/j.ejcts.2010.02.042. [DOI] [PubMed] [Google Scholar]
- 13.Befeler A.S., Palmer D.E., Hoffman M., Longo W., Solomon H., Di Besceglie A.M. The safety of intra-abdominal surgery in patients with cirrhosis: model for end-stage liver disease score is superior to child-turcotte-pugh classification in predicting outcome. Arch Surg. 2005;140:650–654. doi: 10.1001/archsurg.140.7.650. [DOI] [PubMed] [Google Scholar]
- 14.Chen C.C., Hsu P.W., Lee S.T., et al. Brain surgery in patients with liver cirrhosis. J Neurosurg. 2012;117:348–353. doi: 10.3171/2012.4.JNS111338. [DOI] [PubMed] [Google Scholar]
- 15.Nyberg E.M., Batech M., Cheetham T.C., et al. Postoperative risk of hepatic decompensation after orthopedic surgery in patients with cirrhosis. J Clin Transl Hepatol. 2016;4:83–89. doi: 10.14218/JCTH.2015.00049. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Lee J.H., Yu C.S., Lee J.L., et al. Factors affecting the postoperative morbidity and survival of patients with liver cirrhosis following colorectal cancer surgery. Int J Colorectal Dis. 2017;32:521–530. doi: 10.1007/s00384-016-2739-7. [DOI] [PubMed] [Google Scholar]
- 17.Lopez-Delgado J.C., Ballus J., Esteve F., et al. Outcomes of abdominal surgery in patients with liver cirrhosis. World J Gastroenterol. 2016;22:2657–2667. doi: 10.3748/wjg.v22.i9.2657. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Neeff H.P., Streule G.C., Drognitz O., et al. Early mortality and long-term survival after abdominal surgery in patients with liver cirrhosis. Surgery. 2014;155:623–632. doi: 10.1016/j.surg.2013.11.009. [DOI] [PubMed] [Google Scholar]
- 19.Reverter E., Cirera I., Albillos A., et al. The prognostic role of hepatic venous pressure gradient in cirrhotic patients undergoing elective extrahepatic surgery. J Hepatol. 2019;71:942–950. doi: 10.1016/j.jhep.2019.07.007. [DOI] [PubMed] [Google Scholar]
- 20.Fink M.A., Berry S.R., Gow P.J., et al. Risk factors for liver transplantation waiting list mortality. J Gastroenterol Hepatol. 2007;22:119–124. doi: 10.1111/j.1440-1746.2006.04422.x. [DOI] [PubMed] [Google Scholar]
- 21.Garcia Ferraira L., Rezende Anastácio L., Soares Lima A., Davisson Correia MI Touslon. Predictors of mortality in patients on the waiting list for liver transplantation. Nutr Hosp. 2013;28:914–919. doi: 10.3305/nh.2013.28.3.6333. [DOI] [PubMed] [Google Scholar]
- 22.Kim W.R., Therneau T.M., Benson J.T., et al. Deaths on the liver transplantation waiting list: an analysis of competing risks. Hepatology. 2006;43:345–351. doi: 10.1002/hep.21025. [DOI] [PubMed] [Google Scholar]
- 23.Mozzati M., Martinasso G., Cocero N., et al. Superpulsed laser therapy on healing process after tooth extraction in patients waiting for liver transplantation. Laser Med Sci. 2012;27:353–359. doi: 10.1007/s10103-010-0880-z. [DOI] [PubMed] [Google Scholar]
- 24.Merion R.M., Wolfe R.A., Dykstra D.M., Leichtman A.B., Gillespie B., Held P.J. Longitudinal assessment of mortality risk among candidates for liver transplantation. Liver Transplant. 2003;9:12–18. doi: 10.1053/jlts.2003.50009. [DOI] [PubMed] [Google Scholar]
- 25.Dindo D., Dermartines N., Clavien P.A. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240:205–2013. doi: 10.1097/01.sla.0000133083.54934.ae. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.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. doi: 10.1053/j.gastro.2013.02.042. [DOI] [PubMed] [Google Scholar]
- 27.Meunier K., Mucci S., Quentin V., Azoulay R., Arnaud J.P., Hamy A. Colorectal surgery in cirrhotic patients: assessment of operative morbidity and mortality. Dis Colon Rectum. 2008;51:1225–1231. doi: 10.1007/s10350-008-9336-y. [DOI] [PubMed] [Google Scholar]
- 28.Telem D.A., Schiano T., Goldstone R., et al. Factors that predict outcome of abdominal operations in patients with advanced cirrhosis. Clin Gastroenterol Hepatol. 2010;8:451–457. doi: 10.1016/j.cgh.2009.12.015. [DOI] [PubMed] [Google Scholar]
- 29.Wiesner R., Edwards E., Freeman R., et al. Model for end-stage liver disease (MELD) and allocation of donor livers. Gastroenterology. 2003;124:91–96. doi: 10.1053/gast.2003.50016. [DOI] [PubMed] [Google Scholar]
- 30.Zaccherini G., Baldassarre M., Bartoletti M., et al. Prediction of nosocomial acute-on-chronic liver failure in patients with cirrhosis admitted to hospital with acute decompensation. JHEP Reports. 2019;1:270–277. doi: 10.1016/j.jhepr.2019.07.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Trebicka J., Fernandez J., Papp M., et al. The PREDICT study uncovers three clinical courses of acutely decompensated cirrhosis that have distinct pathophysiology. J Hepatol. 2020;73:842–854. doi: 10.1016/j.jhep.2020.06.013. [DOI] [PubMed] [Google Scholar]
- 32.Somsouk M., Kornfield R., Vittinghoff E., Inadomi J.M., Biggins S.W. Moderate ascites identifies patients with low model for end-stage liver disease scores awaiting liver transplantation who have a high mortality risk. Liver Transplant. 2011;17:129–136. doi: 10.1002/lt.22218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Jang H.J., Kim J.H., Song H.H., et al. Clinical outcomes of patients with liver cirrhosis who underwent curative surgery for gastric cancer: a retrospective multi-center study. Dig Dis Sci. 2008;53:399–404. doi: 10.1007/s10620-007-9884-3. [DOI] [PubMed] [Google Scholar]
- 34.Buechter M., Kersting S., Gerken G., Kahraman A. Enzymatic liver function measured by LiMAx – a reliable diagnostic and prognostic tool in chronic liver disease. Sci Rep. 2019;9:1–8. doi: 10.1038/s41598-019-49746-1. [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.