Structured Abstract
Introduction
The International Study Group of Liver Surgery’s (ISGLS) criteria stratifies PHLF into grades A, B, and C. The clinical significance of these grades has not been fully established.
Methods
The American College of Surgeons National Surgical Quality Improvement Program (ACS NSQIP) hepatectomy-targeted database was analyzed. Outcomes between patients without PHLF, with grade A PHLF, and grade B or C PHLF were compared. Univariate and multivariable logistic regression were performed.
Results
6,274 adults undergoing elective major hepatectomy were included in the analysis. The incidence of grade A PHLF was 4.3% and grade B or C was 5.3%. Mortality was similar between patients without PHLF (1.2%) and with grade A PHLF (1.1%), but higher in those with grades B or C PHLF (25.4%). Overall morbidities rates were 19.3%, 41.7%, and 72.8% in patients without PHLF, with grade A PHLF, and with grade B or C PHLF respectively (p<0.001). Grade A PHLF was associated with increased morbidity (grade A: OR 2.7 [95% CI 2.0–3.5]), unplanned reoperation (grade A: OR 3.4 [95% CI 2.2–5.1]), non-operative intervention (grade A: OR 2.6 [95% CI 1.9–3.6]), LOS (grade A: OR 3.1 [95% CI 2.3–4.1]), and readmission (grade A: OR 1.8 [95% CI 1.3–2.5]) compared to patients without PHLF.
Conclusions
Although mortality was similar between patients without PHLF and with grade A PHLF, other postoperative outcomes were notably inferior. Grade A PHLF is a clinically distinct entity with relevant associated postoperative morbidity.
Keywords: hepatectomy, liver failure, colorectal liver metastases, hepatocellular carcinoma
INTRODUCTION
Liver resection remains a mainstay of curative treatment for patients with primary and secondary liver tumors1–5. With any liver resection, the size and quality of remnant liver is associated with the risk of dysfunction and failure postoperatively6–8. The reported incidence of post-hepatectomy liver failure (PHLF) varies from 1–34 percent across the literature9–17. The broad range in incidence can be attributed to the fact that most studies are performed at single centers, focused on a single diagnosis, or both. Patient populations also tend to vary based on the extent of resection as well as baseline liver function, age, comorbidities, and perioperative management9–17.
A significant burden of the postoperative morbidity and mortality after an elective resection is attributed to PHLF6–8,18. Proper perioperative risk assessment is required to anticipate PHLF. In 2011, the International Study Group of Liver Surgery (ISGLS) proposed a consensus definition and grading of PHLF19. The ISGLS criteria were supported by a single center study in 2011 by Rahbari et al.20,21. This study, which included patients undergoing any hepatectomy at a single institution, evaluated the grades as an early postoperative predictor of outcome, and found that patients with PHLF (grades A, B, and C) had significantly higher mortality rates. Subsequent studies evaluating the utility of the ISGLS criteria show mixed results with respect to predicting prognosis, especially mortality, as well as postoperative morbidity. Despite these analyses, the ISGLS stratification includes both objective laboratory values as well as clinical status. This grading system can allow surgeons to understand the relevant consequences that follow PHLF and guide postoperative decision-making.
Although the ISGLS criteria is a more inclusive definition of PHLF, further evaluation of the clinical relevance of PHLF is warranted. The ISGLS criteria for grade A PHLF represents deterioration of postoperative liver function that does not require a change in the patient’s clinical management20. However, whether this definition pertains to patients who share similar postoperative outcomes compared to those without PHLF is not clear20. To better investigate these definitions, this study analyzed a national multi-center database to compare the outcomes across PHLF grades after a major elective hepatectomy. Primarily the study sought to characterize the morbidity and mortality associated specifically with grade A PHLF. Secondarily, the study sought to identify predictors of grade A PHLF.
MATERIALS AND METHODS
Study Design and Data Source
Prior to conducting the study, the STROBE guidelines were reviewed to ensure that proper reporting practices could be observed. After approval from the institutional review board, the ACS National Surgical Quality Improvement Program (NSQIP) was used to identify patients who underwent hepatectomy. The ACS NSQIP database includes clinical and laboratory data from nearly 700 participating hospitals to examine outcomes associated with improved quality of surgical services and patient satisfaction. NSQIP is a validated and prospectively maintained surgical registry. Nurse abstractors are trained to collect over 150 variables from patient records including demographics, preoperative comorbidities, operative characteristics, and postoperative events. The accuracy and reliability of data collection is ensured through periodic auditing of participating hospitals. In addition, ACS NSQIP procedure targeted datasets provide additional procedure-specific variables related to the specific disease sites including malignant histology, staging characteristics, use of neoadjuvant therapy, and margin positivity. The hepatectomy procedure specific targeted database has been maintained by ACS NSQIP since 2014 and includes 30 hepatectomy-specific variables. The PHLF variable mimics the ISGLS definition. Data from the hepatectomy targeted database from 2014 through 2018 were used to identify patients for this study. During the study period, the number of institutions targeting hepatectomy increased from 85 to 140.
Study Cohort
All adult patients who underwent a major elective hepatectomy from 2014 to 2018 were identified in the hepatectomy-targeted file by the following primary Concurrent Procedure Terminology (CPT) codes: 47122 (trisegmentectomy), 47125 (total left hepatectomy), and 47130 (total right hepatectomy). Patients missing American Society of Anesthesiologists (ASA) physical status classification, undergoing emergent hepatectomy, and undergoing non-major hepatectomy (<3 segments) were excluded. Patients with ASA V were also excluded. Patients undergoing non-major hepatectomy were excluded to maximize the ability to characterize the clinical relevance of grade A PHLF.
Demographic and preoperative factors included age, race, gender, ASA classification, body mass index (BMI) class, smoking status, comorbidities (diabetes mellitus, disseminated cancer, bleeding disorders, ascites, hypertension, and weight loss >10 percent in 6 months), laboratory values (albumin, bilirubin, and AST). Continuous variables such as age, BMI, and laboratory values were reclassified as normal and abnormal based on commonly accepted reference ranges. Operative variables of interest included type of resection, operative approach (minimally invasive approach or open), biliary reconstruction method, use of Pringle maneuver, concurrently performed colorectal procedures, biliary stent placement, and operative drain placement. For this study, any reported concurrent colectomy or proctectomy was considered a concurrent colorectal procedure. Viral hepatitis status, completion of neoadjuvant therapy, liver texture, and final pathology were also considered in the analysis. The hepatectomy-targeted data set defines liver texture as cirrhotic, fibrotic, congested, fatty, or normal. This was reclassified as normal or abnormal, with cirrhotic, fibrotic, congested, and fatty being considered abnormal. Any binary variables reported as unknown were grouped with the most common category or categorized as “no.” Any unknown laboratory values were considered normal.
Exposure Classification and Outcomes
The ISGLS criteria were used to define PHLF. The ISGLS considers PHLF a postoperatively acquired deterioration in the synthetic, excretory, and detoxifying functions of the liver characterized by an increased INR (or need of clotting factors) and hyperbilirubinemia on or after postoperative day five. This deterioration is graded based on severity (Table 1).
Table 1.
Definition and Severity Grading of Post-Hepatectomy Liver Failure (PHLF) based on the International Study Group of Liver Surgery (ISGLS) Criteria.
| Definition | A postoperatively acquired deterioration in the ability of the liver (in patients with normal and abnormal liver function) to maintain its synthetic, excretory, and detoxifying functions, characterized by an increased INR (or need of clotting factors to maintain normal INR) and hyperbilirubinemia (according to the normal cut-off levels defined by the local laboratory) on or after postoperative day 5. If INR or serum bilirubin concentration is increased preoperatively, PHLF is defined by an increasing INR (decreasing prothrombin time) and increasing serum bilirubin concentration on or after postoperative day 5 (compared with the values of the previous day). Other obvious causes for the observed biochemical and clinical alterations such as biliary obstruction should be ruled out. |
| Grade | |
| A | PHLF resulting in abnormal laboratory parameters but requiring no change in the clinical management of the patient |
| B | PHLF resulting in a deviation from the regular clinical management but manageable without invasive treatment. |
| C | PHLF resulting in a deviation from the regular clinical management and requiring invasive treatment. |
The primary outcomes of interest included overall morbidity and postoperative length of stay (LOS). Secondary outcomes of interest included unplanned reoperation, invasive nonoperative intervention, readmission, and 30-day mortality. Any patient who fell in the fourth quartile LOS was included in a “prolonged stay” category for regression models. Overall morbidity is a composite variable based on the ACS NSQIP hepatopancreatobiliary (HPB) Collaborative. The ACS NSQIP HPB Collaborative was established in 2013 to reduce variation in care delivery and to improve outcomes of patients undergoing HPB procedures. Overall morbidity is a measure utilized by the collaborative. Overall morbidity includes one or more of the following but excludes death: superficial incisional surgical site infection (SSI), deep incisional SSI, organ/deep space SSI, wound disruption, pneumonia, unplanned intubation, ventilator time >48 hours, progressive renal insufficiency, acute renal failure, urinary tract infection, stroke/cerebral vascular accident, cardiac arrest, myocardial infarction, or sepsis. Analogous composite scores have been used and shown to be reliable performance measures in surgery. The “any-or-none” outcome measure (optimal hepatic surgery) is also consistent with other outcome measures that are endorsed by the National Quality Forum.
Statistical Analysis
Chi-squared and Kruskal-Wallis tests were used to compare patient and operative characteristics between patients without PHLF, grade A PHLF, and grades B or C PHLF. The association between PHLF severity and postoperative outcomes were further evaluated by multivariable logistic regression. Due to variation in the number of outcome events, regression models were constructed separately for each outcome to reduce the risk for overfitting. Predictor selection was based on clinical relevance established by previous literature and constructed to include covariates with p<0.1 on univariate analysis. Non-contributory predictors were removed through a backward stepwise selection. Presented odds ratios were the results of the multivariable logistic regression models. Reported confidence intervals were to a 95 percent level of significance. Tests of significance were two-sided with a threshold of p<0.05. All statistical analyses were performed using STATA version 14.2 (StataCorp, College Station, TX).
RESULTS
Study Cohort
Of the 7,656 patients who met the inclusion criteria, 6,274 patients were included in the analysis after applying the exclusion criteria. A total of 602 were diagnosed with PHLF (9.6 percent) after a major hepatectomy. Two hundred seventy-one patients were identified to have grade A (4.3 percent overall) and 331 patients were found to have grades B or C (5.3 percent overall). The majority of patients underwent an open resection (N=5,625, 89.7 percent) primarily for a secondary (metastatic) tumor (N=2,786, 44.4 percent). The plurality of patients had a total right hepatectomy (N=3,061 48.8 percent) and a small proportion had a concurrent colorectal procedure (N=234, 3.7 percent). Additional patient-level clinical, operative, and pathologic characteristics associated with PHLF severity are listed in Table 2.
Table 2.
Patient demographics and characteristics by PHLF grade
| None, (N=5672) | Grade A, (N=271) | Grade B/C, (N=331) | P-Value | |
|---|---|---|---|---|
| ASA Classification | ||||
| 1 and 2 | 1574 (27.8) | 76 (28) | 55 (16.6) | <0.001 |
| 3 and 4 | 4098 (72.3) | 195 (72) | 276 (83.4) | |
| Gender | ||||
| Female | 2871 (50.6) | 92 (34) | 131 (39.6) | <0.001 |
| Male | 2801 (49.4) | 179 (66.1) | 200 (60.4) | |
| BMI Class | ||||
| Underweight (<18.5) | 102 (1.8) | 4 (1.5) | 9 (2.7) | 0.476 |
| Normal (18.5–24.9) | 1863 (32.9) | 93 (34.3) | 117 (35.4) | |
| Overweight (25.0–29.9) | 1947 (34.3) | 96 (35.4) | 97 (29.3) | |
| Obese (≥30) | 1760 (31) | 78 (28.8) | 108 (32.6) | |
| Age | ||||
| <55 | 2023 (35.7) | 90 (33.2) | 77 (23.3) | <0.001 |
| 55–64 | 1560 (27.5) | 66 (24.4) | 83 (25.1) | |
| ≥65 | 2089 (36.8) | 115 (42.4) | 171 (51.7) | |
| Race | ||||
| White | 3394 (59.8) | 134 (49.5) | 192 (58) | <0.001 |
| Non-Hispanic Black | 436 (7.7) | 19 (7) | 17 (5.1) | |
| Hispanic | 270 (4.8) | 10 (3.7) | 11 (3.3) | |
| Other | 1572 (27.7) | 108 (39.9) | 111 (33.5) | |
| Diabetes Mellitus | ||||
| No | 4775 (84.2) | 218 (80.4) | 246 (74.3) | <0.001 |
| Yes | 897 (15.8) | 53 (19.6) | 85 (25.7) | |
| Disseminated Cancer | ||||
| No | 3421 (60.3) | 167 (61.6) | 205 (61.9) | 0.777 |
| Yes | 2251 (39.7) | 104 (38.4) | 126 (38.1) | |
| Ascites | ||||
| No | 5653 (99.7) | 270 (99.6) | 327 (98.8) | 0.044 |
| Yes | 19 (0.3) | 1 (0.4) | 4 (1.2) | |
| Hypertension | ||||
| No | 3218 (56.7) | 149 (55) | 169 (51.1) | 0.115 |
| Yes | 2454 (43.3) | 122 (45) | 162 (48.9) | |
| Weight loss (>10% in 6 months) | ||||
| No | 5379 (94.8) | 256 (94.5) | 296 (89.4) | <0.001 |
| Yes | 293 (5.2) | 15 (5.5) | 35 (10.6) | |
| Pre-operative Albumin | ||||
| <3.4 | 541 (9.5) | 23 (8.5) | 71 (21.5) | <0.001 |
| >3.4 or unknown | 5131 (90.5) | 248 (91.5) | 260 (78.6) | |
| Pre-operative AST | ||||
| <40 or unknown | 4347 (76.6) | 181 (66.8) | 189 (57.1) | <0.001 |
| >40 | 1325 (23.4) | 90 (33.2) | 142 (42.9) | |
| Pre-operative Bilirubin | ||||
| <1.2 or unknown | 5191 (91.5) | 248 (91.5) | 271 (81.9) | <0.001 |
| >1.2 | 481 (8.5) | 23 (8.5) | 60 (18.1) | |
| Smoker | ||||
| No | 4751 (83.8) | 229 (84.5) | 268 (81) | 0.379 |
| Yes | 921 (16.2) | 42 (15.5) | 63 (19) | |
| Bleeding disorders | ||||
| No | 5529 (97.5) | 262 (96.7) | 312 (94.3) | 0.002 |
| Yes | 143 (2.5) | 9 (3.3) | 19 (5.7) | |
| Pathology | ||||
| Benign | 952 (16.8) | 19 (7) | 21 (6.3) | <0.001 |
| Primary | 1909 (33.7) | 119 (43.9) | 186 (56.2) | |
| Secondary or unknown | 2811 (49.6) | 133 (49.1) | 124 (37.5) | |
| Liver Texture | ||||
| Abnormal | 1265 (22.3) | 79 (29.2) | 103 (31.1) | <0.001 |
| Normal or Unknown | 4407 (77.7) | 192 (70.9) | 228 (68.9) | |
| Viral Hepatitis | ||||
| No or Unknown | 5079 (89.6) | 234 (86.4) | 287 (86.7) | 0.077 |
| Yes | 593 (10.5) | 37 (13.7) | 44 (13.3) | |
| Neoadjuvant Therapy | ||||
| No or Unknown | 3626 (63.9) | 165 (60.9) | 197 (59.5) | 0.173 |
| Yes | 2046 (36.1) | 106 (39.1) | 134 (40.5) | |
| Extent of resection | ||||
| Total Left | 1660 (29.3) | 27 (10) | 27 (8.2) | <0.001 |
| Total Right | 2707 (47.7) | 176 (64.9) | 178 (53.8) | |
| Trisegmentectomy | 1305 (23) | 68 (25.1) | 126 (38.1) | |
| Operative Approach | ||||
| MIS | 623 (11) | 10 (3.7) | 16 (4.8) | <0.001 |
| Open | 5049 (89) | 261 (96.3) | 315 (95.2) | |
| Billiary Reconstruction | ||||
| No | 5020 (88.5) | 224 (82.7) | 217 (65.6) | <0.001 |
| Yes | 652 (11.5) | 47 (17.3) | 114 (34.4) | |
| Pringle Maneuver | ||||
| No | 4082 (72) | 192 (70.9) | 223 (67.4) | 0.187 |
| Yes | 1590 (28) | 79 (29.2) | 108 (32.6) | |
| Concurrent Colorectal Procedure | ||||
| Yes | 199 (3.5) | 16 (5.9) | 19 (5.7) | 0.018 |
| No | 5473 (96.5) | 255 (94.1) | 312 (94.3) | |
| Biliary Stent Placement | ||||
| No | 5142 (90.7) | 237 (87.5) | 255 (77) | <0.001 |
| Yes | 530 (9.3) | 34 (12.6) | 76 (23) | |
| Drain Placement | ||||
| No | 2474 (43.6) | 113 (41.7) | 105 (31.7) | <0.001 |
| Yes | 3198 (56.4) | 158 (58.3) | 226 (68.3) |
Incidence of Postoperative Outcomes
The 30-day mortality rate in the patient cohort was 2.5 percent overall, 1.2 percent for those without PHLF, and 14.4 percent for those with PHLF. Mortality was similar between patients without PHLF (1.2 percent) and with grade A (1.1 percent), but higher in those with grades B or C (25.4 percent) (Table 3). Overall median postoperative LOS was 6 days (IQR 5–8), 9 days (IQR 6–13) for grade A, and 15 days (IQR 9–25) for grades B or C (p<0.001, Figure 1). For the entire patient cohort after major hepatectomy, the rate of overall morbidity was 23.1 percent non-operative interventions 12.5 percent, unplanned reoperations 4.4 percent, and readmissions 12.5 percent. Except for the rate of readmissions, the incidence of all postoperative outcomes was higher in patients with PHLF and increased with greater severity of PHLF (p<0.001). More specifically, the incidence of overall morbidity was 41.7 percent in grade A and 72.8 percent in grades B or C patients (p<0.001). Grade A patients also had fewer postoperative interventions (non-operative 25.8 percent; unplanned reoperation 10.7 percent) compared to patients with grades B or C PHLF (non-operative 49.6 percent; unplanned reoperation 21.8 percent) (p<0.001). Readmission rates were similar between grade A (21.0 percent) and grades B or C (20.5 percent), but greater than those without PHLF (11.6 percent) (p<0.001). The results of the incidence of postoperative outcomes based on PHLF severity are listed in Table 3.
Table 3.
Incidence of Post-Operative Outcomes
| None, No. (%) | Grade A, No. (%) | Grade B/C, No. (%) | P-Value | |
|---|---|---|---|---|
| 30-day Mortality | 69 (1.2) | 3 (1.1) | 84 (25.4) | <0.001 |
| Overall Morbidity | 1095 (19.3) | 113 (41.7) | 241 (72.8) | <0.001 |
| Readmission | 657 (11.6) | 57 (21) | 68 (20.5) | <0.001 |
| Non-operative intervention | 607 (10.7) | 70 (25.8) | 164 (49.6) | <0.001 |
| Unplanned Reoperation | 176 (3.1) | 29 (10.7) | 72 (21.8) | <0.001 |
Figure 1.
Hospital Length of Stay by PHLF grade.
Multivariable Regression
The adjusted odds of 30-day mortality was not significantly different between grade A patients and those without PHLF (grade A OR: 0.66 [95% CI: 0.21–2.1]). Any severity of PHLF was associated with increased odds of overall 30-day morbidity, serious morbidity, need for a non-operative invasive procedure, unplanned reoperation, increased postoperative LOS, and readmission. Patients with grade A PHLF had greater odds of readmission compared to patients without PHLF and those with grades B or C (grade A: OR 1.8 [95% CI 1.3–2.5], grade B or C: OR 1.4 [95% CI 1.0–1.9]). We also found that with increased PHLF severity, patients had an increased odds of overall morbidity (grade A: OR 2.7 [95% CI 2.0–3.5]; grade B or C: OR 7.7 [95% CI 5.8–10.1]) and increased odds of prolonged LOS (grade A: OR 3.1 [95% CI 2.3–4.1]; grade B or C: OR 5.9 [95% CI 4.3–8.3]). Results from the adjusted multivariable models are depicted in Figure 2.
Figure 2.
Adjusted Odds-Ratios of Post-Operative Outcomes by PHLF Grade
DISCUSSION
The ISGLS criteria has been embraced as a means to clinically describe post-hepatectomy liver failure (PHLF). This analysis characterized the risk factors and the postoperative outcomes of grade A PHLF. Furthermore, patients without PHLF as well as more severe PHLF were compared. Patients with grade A PHLF had similar demographic and preoperative characteristics to those without PHLF and with grades B or C. Additionally, with the exception of mortality, patients with grade A had inferior outcomes compared to patients without PHLF. However, grade A patients had lower overall morbidity, fewer non-operative interventions, and fewer unplanned reoperations than grade B or C patients.
Among the analyzed cohort, the incidence of PHLF was 9.6 percent after major hepatectomy. The incidence of grade A PHLF was 4.3 percent and the incidence of grades B or C was 5.3 percent based on the ISGLS criteria. These results are below the incidence rate reported by the seminal study done by Rahbari et al. to validate the ISGLS criteria, reporting an incidence of 12 percent21. Given the patient cohort excluded those who underwent minor hepatectomies, the expected incidence of PHLF in the current cohort should reasonably exceed the findings reported by Rahbari et al. However, the current findings may be more representative of all patients undergoing hepatectomy, given the large, heterogeneous nature of the study population. Altogether, the analysis confirms that PHLF is not an uncommon complication in those undergoing a major liver resection.
The current analysis confirms Rahbari et al.’s strong association grade B or C PHLF with mortality21. Skrzypczyk et al. reported similar findings when comparing the ISGLS criteria to other commonly used PHLF definitions such as the “50–50 criteria” and peak bilirubin >7mg/dL22. They reported a mortality incidence of 0 percent in grade A, 23 percent in grade B, and 54 percent in grade C22. These rates are concordant with the results in this study of 1.1 percent in grade A and 25.4 percent in grades B or C.
The likelihood of developing a major postoperative complication, as defined by ACS-NSQIP composite scores, was higher with increasing PHLF severity. Major postoperative complications are measured differently across studies that define morbidity based on the Clavien-Dindo Classification, the ISGLS definition, or a list of common postoperative complications23. Fukushima et al., in their single-center study of 210 patients who underwent a resection for hepatocellular carcinoma found postoperative complication rates defined by the ISGLS criteria, to be 18.7 percent in patients without PHLF, 10 percent in grade A PHLF, 58 percent in grade B PHLF, and 100 percent in grade C PHLF24. Other studies, show a 16.5 to 27 percent incidence of major morbidity defined as Clavien-Dindo III-V17,24,25. The current study reports a higher percentage of patients with complications that could be classified as Clavien-Dindo grade III-V. For example, the rate of invasive non-operative intervention was 25.8 percent in grade A and 49.6 percent in grade B or C PHLF. Similarly, the rate of unplanned reoperation was 10.7 percent in grade A and 21.8 percent in grade B or C PHLF. Again, these rates are somewhat higher than the Clavien-Dindo grade III-V morbidity rates reported previously. Finally, the current findings confirm the known characteristics and peri-operative risk factors of PHLF, which are supported in the literature: advanced age, male sex, obesity, comorbidities, pre-existing liver disease, operative approach, extent of resection, and biliary reconstruction method6,10,15,19,26.
Most notably, patients with grade A PHLF in the NSQIP cohort had demographic, disease-specific, and intraoperative characteristics that fell between the spectrum of patients without PHLF and those with more severe PHLF. The ISGLS criteria defines grade A PHLF as any changes in laboratory values without notable organ dysfunction requiring invasive interventions20. Compared to NSQIP patients without PHLF, grade A patients had a significantly higher incidence of postoperative complications, rate of invasive intervention, postoperative LOS, and rate of readmission. These results support the conclusion that grade A PHLF is a distinct and clinically relevant designation. It can therefore be applied, as appropriate, to patients undergoing hepatectomy and be used to guide management.
There are a few limitations to this study. Results from the ACS NSQIP procedure targeted hepatectomy database are obtained from hospitals that have opted-in to share their patient level data, introducing a small amount of ascertainment bias. Also, hospital-level data are lacking, making it difficult to adjust for center-specific level variation in PHLF and related outcomes. As with any large database, the variables which are collected and their validity are subject to scrutiny. Specifically, pre- and postoperative liver volumes are not collected in the hepatectomy-target file. However, given the prospective and validated nature of NSQIP, these data are like some of the largest and highest quality available given this investigative setting.
In addition, the ACS NSQIP database only reports outcomes within 30 days after surgery27. Given the follow up, it is possible that the reported rates and odds of readmission are underestimated. Many PHLF patients in the current cohort had prolonged initial postoperative stays beyond the 30-day period captured by the database. This may explain the increased readmission rate for patients without PHLF or increased odds of readmission for patients with grade A PHLF. It could, alternatively, be that the increased LOS associated with increasing PHLF grade leads to the identification and rectification of factors resulting in readmission. As the LOS is shorter in patients without PHLF, these same factors leading to readmission may not be apparent until after discharge, resulting in the readmission event itself. Notwithstanding, increased morbidity and mortality have been reported in the literature 90-days postoperatively in patients with PHLF. Other studies also report that 90-day mortality can be up to 47 percent higher than 30-day mortality14,24,25,28. It would be prudent for future investigations to include longer follow up times to further characterize the natural history of patients with and without PHLF.
CONCLUSION
Grade A PHLF is a clinically relevant entity in patients undergoing elective major hepatectomy. Postoperative complications, LOS, and readmission rate were inferior in patients who experienced grade A PHLF. Establishing outcomes associated with each grade of PHLF can be used to enhance preoperative patient selection and guide management in the postoperative setting.
Synopsis:
A correction of a sentence in the third paragraph of the introduction for clarity was made. Additionally, a sentence at the beginning of the results was changed for clarity.
Conflicts of Interest and Sources of Funding
The authors have no pertinent conflicts of interest to disclose.
Research reported in this publication was supported, in part, by the National Institutes of Health’s training grant 5R38CA245095; Surgical Multispecialty Access to Research in Residency Training (SMART) at Northwestern University and the Steven J. Stryker, MD Gastrointestinal Surgery Research and Education Endowment. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.
References
- 1.Benson AB, D’Angelica MI, Abbott DE, et al. Hepatobiliary Cancers, Version 1.2017. J Natl Compr Canc Netw. 2017;15:563–573. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Ebata T, Mizuno T, Yokoyama Y, et al. Surgical resection for Bismuth type IV perihilar cholangiocarcinoma. Br J Surg. 2018;105:829–838. [DOI] [PubMed] [Google Scholar]
- 3.Jones RP, Jackson R, Dunne DFJ, et al. Systematic review and meta-analysis of follow-up after hepatectomy for colorectal liver metastases2. British Journal of Surgery. 2012;99:477–486. [DOI] [PubMed] [Google Scholar]
- 4.Merath K, Chen Q, Bagante F, et al. Textbook Outcomes Among Medicare Patients Undergoing Hepatopancreatic Surgery. Annals of Surgery. 2020;271:1116. [DOI] [PubMed] [Google Scholar]
- 5.Spolverato G, Kim Y, Alexandrescu S, et al. Is Hepatic Resection for Large or Multifocal Intrahepatic Cholangiocarcinoma Justified? Results from a Multi-Institutional Collaboration. Ann Surg Oncol. 2015;22:2218–2225. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Lafaro K, Buettner S, Maqsood H, et al. Defining Post Hepatectomy Liver Insufficiency: Where do We stand? J Gastrointest Surg. 2015;19:2079–2092. [DOI] [PubMed] [Google Scholar]
- 7.Rahnemai-Azar AA, Cloyd JM, Weber SM, et al. Update on Liver Failure Following Hepatic Resection: Strategies for Prediction and Avoidance of Post-operative Liver Insufficiency. J Clin Transl Hepatol. 2018;6:97–104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.van den Broek MAJ, van Dam RM, van Breukelen GJP, et al. Development of a composite endpoint for randomized controlled trials in liver surgery. British Journal of Surgery. 2011;98:1138–1145. [DOI] [PubMed] [Google Scholar]
- 9.Alizai PH, Haelsig A, Bruners P, et al. Impact of liver volume and liver function on posthepatectomy liver failure after portal vein embolization– A multivariable cohort analysis. Ann Med Surg (Lond). 2017;25:6–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Bagante F, Tsilimigras DI, Pawlik TM. ASO Author Reflections: Minimally Invasive Surgery for Hepatocellular Carcinoma in the Setting of Portal Vein Hypertension. Ann Surg Oncol. 2020;27:3372–3373. [DOI] [PubMed] [Google Scholar]
- 11.Bagante F, Ruzzenente A, Beal EW, et al. Complications after liver surgery: a benchmark analysis. HPB. 2019;21:1139–1149. [DOI] [PubMed] [Google Scholar]
- 12.Birgin E, Tesfazgi W, Knoth M, et al. Evaluation of the New ISGLS Definitions of Typical Posthepatectomy Complications. Scand J Surg. 2019;108:130–136. [DOI] [PubMed] [Google Scholar]
- 13.Filicori F, Keutgen XM, Zanello M, et al. Prognostic criteria for postoperative mortality in 170 patients undergoing major right hepatectomy. Hepatobiliary & Pancreatic Diseases International. 2012;11:507–512. [DOI] [PubMed] [Google Scholar]
- 14.Gilg S, Sandström P, Rizell M, et al. The impact of post-hepatectomy liver failure on mortality: a population-based study. Scandinavian Journal of Gastroenterology. 2018;53:1335–1339. [DOI] [PubMed] [Google Scholar]
- 15.Li J, Moustafa M, Freiwald-Bibiza E, et al. Is It Feasible to Standardize a Composite Postoperative Complication Reporting System for Liver Resection? J Gastrointest Surg. 2020;24:2748–2755. [DOI] [PubMed] [Google Scholar]
- 16.Van Den Broek MAJ, Olde Damink SWM, Dejong CHC, et al. Liver failure after partial hepatic resection: definition, pathophysiology, risk factors and treatment. Liver International. 2008;28:767–780. [DOI] [PubMed] [Google Scholar]
- 17.Zheng Y, Yang H, He L, et al. Reassessment of different criteria for diagnosing post-hepatectomy liver failure: a single-center study of 1683 hepatectomy. Oncotarget. 2017;8:89269–89277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Schroeder RA, Marroquin CE, Bute BP, et al. Predictive Indices of Morbidity and Mortality After Liver Resection. Ann Surg. 2006;243:373–379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Yadav K, Shrikhande S, Goel M. Post Hepatectomy Liver Failure: Concept of Management. J Gastrointest Canc. 2014;45:405–413. [DOI] [PubMed] [Google Scholar]
- 20.Rahbari NN, Garden OJ, Padbury 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]
- 21.Rahbari NN, Reissfelder C, Koch M, et al. The Predictive Value of Postoperative Clinical Risk Scores for Outcome After Hepatic Resection: A Validation Analysis in 807 Patients. Ann Surg Oncol. 2011;18:3640–3649. [DOI] [PubMed] [Google Scholar]
- 22.Skrzypczyk C, Truant S, Duhamel A, et al. Relevance of the ISGLS Definition of Posthepatectomy Liver Failure in Early Prediction of Poor Outcome After Liver Resection: Study on 680 Hepatectomies. Annals of Surgery. 2014;260:865. [DOI] [PubMed] [Google Scholar]
- 23.Kuramitsu K, Fukumoto T, Kobayashi T, et al. The Incidence of Posthepatectomy Liver Failure Defined by the International Study Group of Liver Surgery among Living Donors. J Gastrointest Surg. 2016;20:757–764. [DOI] [PubMed] [Google Scholar]
- 24.Fukushima K, Fukumoto T, Kuramitsu K, et al. Assessment of ISGLS Definition of Posthepatectomy Liver Failure and Its Effect on Outcome in Patients with Hepatocellular Carcinoma. J Gastrointest Surg. 2014;18:729–736. [DOI] [PubMed] [Google Scholar]
- 25.Vibert E, Pittau G, Gelli M, et al. Actual incidence and long-term consequences of posthepatectomy liver failure after hepatectomy for colorectal liver metastases. Surgery. 2014;155:94–105. [DOI] [PubMed] [Google Scholar]
- 26.Wang H, Lu S-C, He L, et al. A study on risk factors and diagnostic efficiency of posthepatectomy liver failure in the nonobstructive jaundice. Medicine (Baltimore). 2018;97:e9963. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Hall BL, Hamilton BH, Richards K, et al. Does Surgical Quality Improve in the American College of Surgeons National Surgical Quality Improvement Program: An Evaluation of All Participating Hospitals. Annals of Surgery. 2009;250:363. [DOI] [PubMed] [Google Scholar]
- 28.Hyder O, Pulitano C, Firoozmand A, et al. A Risk Model to Predict 90-Day Mortality among Patients Undergoing Hepatic Resection. J Am Coll Surg. 2013;216:1049–1056. [DOI] [PMC free article] [PubMed] [Google Scholar]