Abstract
Background/Aim: The rate of postoperative morbidity after major hepatobiliary pancreatic (HBP) surgery is estimated to be high. We hypothesized that the frailty and insufficient blood supply of organs caused by arteriosclerosis may lead to increased severity of postoperative complications. The aim of this study was to clarify the relationship between abdominal aortic calcification (AAC) and severity of major HBP surgery postoperative complications. Patients and Methods: A total of 322 major HBP surgeries were performed in Hiroshima University Hospital, Japan from January 2010 to March 2018. The records of 214 patients were retrospectively analyzed to include those with hepatectomy in two or more segments. These included donor hepatectomy, hepatectomy with biliary tract reconstruction, and pancreaticoduodenectomy. We compared the baseline AAC levels (cut-off; 1,000 mm3) with their corresponding levels after HBP surgery complications developed. The complication severity was assessed using the Comprehensive Complication Index (CCI). Results: The high-AAC (N=71) group had significantly increased incidence rates of surgical site infections and a higher CCI, even after propensity matching. Multivariate analysis revealed that a higher AAC was an independent risk factor of severe postoperative complications (CCI ≥40) (OR=10.21, p<0.01). Conclusion: An increased AAC has a corresponding increase in postoperative complication severity after major HBP surgery. Careful anastomosis and infection control are required to prevent the progression of severe postoperative complications among high-AAC patients.
The high risk of postoperative complications after major invasive hepatobiliary pancreatic (HBP) surgeries such as major hepatectomy, pancreaticoduodenectomy (PD), and hepatectomy with biliary reconstruction, is a clinical reality. The morbidity rates of HBP surgery have been reported to range from 40% to 75% (1-5). There have been many attempts in the past to prevent the occurrence of postoperative complications (6). We hypothesized that the frailty and insufficient blood supply of organs caused by arteriosclerosis leads to increased severity of postoperative complications. The relationship between arteriosclerosis and postoperative complications following HBP surgery has not been investigated in previous reports. Our first study reported how abdominal aortic calcifications (AAC), visualized on computed tomography (CT) scans, could predict poor prognosis (7) and biliary complications (8) after liver transplantation. The aim of the present study was to clarify the relationship between AAC and the severity of major HBP surgery postoperative complications.
Patients and Methods
Patient enrollment. A total of 322 major HBP surgeries were performed in Hiroshima University Hospital, Japan from January 2010 to March 2018. Of these, 108 liver transplantations were excluded and only 214 hepatectomies were included in the study, including donor hepatectomy for liver transplantation. Data regarding patient characteristics at the time of surgery (age, sex, primary disease, surgical procedure, body mass index [BMI], diabetes, preoperative biliary drainage, and medical history of cholangitis) were collected from electronic records. We also noted the patient’s surgical factors (operation time and volume of blood loss) along with the preoperative levels of total bilirubin (T-bil), alkaline phosphatase (ALP), alanine aminotransferase (ALT), albumin, international normalized ratio of prothrombin time (PT-INR), white blood cell (WBC), platelets, and C-reactive protein. The study protocol was approved by the Ethics Committee of Hiroshima University (E-1410). The study conforms to the provisions of the Declaration of Helsinki. The need for written informed consent was waived owing to the retrospective nature of the study. The opt-out method to obtain patient consent was utilized at our institution.
HBP surgery. Major HBP surgery was defined as either a pancreaticoduodenectomy (PD), hemihepatectomy or greater, and a hepatopancreaticoduodenectomy. Whole liver transplants were exempt from the study (9). Donor hepatectomies for living donor liver transplantation were included.
Comprehensive complication index (CCI). All complications were evaluated using the Clavien–Dindo classification (10). According to this system, the CCI was reported as an accurate index that reflected the severity of all postoperative complications (11) and summarized the development of complications (12). Severe morbidity was defined as a CCI ≥40 due to massive intraoperative blood loss and combined PD among major hepatectomies for biliary cancer (13).
Evaluation of abdominal aortic calcification (AAC). CT angiographies were performed using a standardized examination protocol on a 320-detector row CT scanner (Aquilion ONE ViSION, Toshiba Medical Systems, Japan). AAC was evaluated using the AZE Virtual Place Lexus64 Anatomia software (AZE Inc., Tokyo, Japan). The AAC volume was automatically calculated for calcifications located in the abdominal aorta (from the origin of the renal artery to the iliac bifurcation) with attenuations greater than the pre-defined level of 130 Hounsfield units (14). The AAC level cut-off (1,000 mm3) was decided using receiver operating characteristic (ROC) curves of high CCI.
Statistical analysis. The JMP statistical software (JMP® 14; SAS Institute Inc., Cary, NC, USA) was used for all statistical analyses. The Pearson’s Chi-squared test was used to determine the significance of differences between dichotomous groups. A one-to-one matching analysis was used on the AAC levels to minimize bias among patients. Propensity scores were estimated using a logistic model which included the following variables: age, sex, surgical procedure, BMI, diabetes, preoperative biliary drainage, medical history of cholangitis, surgical factors (operation time, volume of blood loss), and preoperative blood tests. To adjust for differences in baseline characteristics, one-to-one propensity score models were constructed based on each patient’s estimated propensity score. All variables were included in the multivariate models and the backward elimination method with removal criterion p=0.05 was used to select covariates.
Results
Patient characteristics. We analyzed the records of 164 males and 50 females who had an overall mean age of 55.9±18.0 years at operation. The distribution of patient characteristics is shown in Table I. The underlying diagnosis was liver donation in 87 cases (40.7%), hepatocellular carcinoma (HCC) in 67 cases (31.3%), pancreatic cancer in 8 cases (3.7%), duodenal cancer in 12 cases (5.6%), and cholangiocarcinoma in 27 cases (12.6%). We first divided the patients into two groups according to the AAC levels – high and low (Table II). The high-AAC group comprised a significantly increased number of males and elderly patients. This group also had a greater volume of blood loss.
Patient characteristics.
Background data according to abdominal aortic calcification (AAC) levels prior to propensity matching.
Clinical outcomes of patients before and after the propensity model. The patient characteristics and surgical procedures after propensity score matching are summarized in Table III. The short-term outcomes based on AAC levels are shown in Table IV. Before propensity matching, the high-AAC group had significantly higher rates of surgical site infections (SSI), blood stream infections (BSI), hypotension, re-admissions into an intensive care unit (ICU), and severe postoperative complications (CCI ≥40) compared to the low-AAC group. The high-AAC group had a significantly higher CCI than the low-AAC group (median: 0% vs. 29.6%, p<0.01) (Figure 1A). Table IV also shows the short-term outcomes based on AAC levels after propensity matching. The high-AAC group also had significantly higher rates of SSI. In the propensity score-matched cohort, the high-AAC group still had a significantly higher CCI compared to the low-AAC group. (Median: 0% vs. 23.5%, p<0.01) (Figure 1B).
Background data according to abdominal aortic calcification (AAC) levels after propensity matching.
Short-term outcomes based on abdominal aortic calcification (AAC) levels.
Comprehensive complication index (CCI) according to abdominal aortic calcification (AAC) levels. Before (A) and after (B) propensity score matching.
Risk factors for severe complications. Table V shows the risk factors for severe postoperative complications (CCI ≥40, N=22). The logistic regression model used for stepwise multivariate analysis revealed that the independent risk factors related to a high CCI (≥40) were higher AAC >1,000 mm3 (OR=10.21, 95% CI=2.95-35.25, p<0.01), higher ALT levels >40 U/l (OR=4.64, 95% CI=1.56-13.7, p<0.01), and presence of a biliary tract reconstruction (OR=7.80, 95% CI=2.58-23.60, p<0.01).
Risk factors for high Comprehensive Complication Index (CCI).
Discussion
In this study, we demonstrated that the AAC had a strong relationship with the development of severe complications within 30 days after major HBP surgery. The high-AAC group was also significantly associated with higher SSI and a correspondingly increased CCI, even after propensity matching. In gastroenterological surgery, the AAC was first reported to be an independent risk factor for postoperative pancreatic fistulas among aged patients undergoing PD (15). Aortic calcification was also a risk factor associated with anastomotic leakage after anterior resections in rectal cancer (16) and colorectal surgery (17). However, no other studies on the relationship between AAC and postoperative complications, especially after major HBP surgery, exist. Interestingly, each complication did not significantly increase, even after the propensity matching, despite a high AAC. In contrast, the CCI increased significantly. This result meant that patients with high AAC had a potential risk of developing a worse condition once postoperative complications occurred.
The evaluation of AAC on CT scans was a simple method to assess vascular calcification and severe arteriosclerosis (14). AAC had been associated with coronary artery disease and stroke in the general patient population (18-20), especially those undergoing dialysis (21, 22). In a previous report, we discussed the clinical implications of incidental AAC findings in liver transplantation in connection with post-transplantation survival (7). According to a recent study, a higher intake of dietary zinc had the potential to prevent severe AAC (23). However, the mechanisms of AAC negatively influencing postoperative complications remained unclear (15). Severe AAC was reported to have an association with worse 90-day mortality among sepsis patients (24). The results of this study indicated high risk of SSI, BSI, and shock associated with an increased AAC. It was likely that the AAC was related to the frailty of patients, especially in those with infectious diseases. Advanced aortic calcification could decrease blood flow and develop organ damage under the high invasion of major HBP surgery (24). Though our study did not report any significant difference in bile leakage and pancreatic fistula formation after propensity matching, the high-AAC group had high incidence rates of both complications. AAC could also possibly reflect a variety of underlying atheromatous diseases. It could, therefore, be a surrogate marker for impaired vascularity and poor tissue perfusion (16).
Higher age has also been well known as a strong factor for the progression of AAC. In this study, the number of older patients in the high-AAC group was higher than that in the low-AAC group. It must be emphasized that surgeons need to identify high-risk populations among elderly patients. In terms of hepatectomy, a higher incidence of pulmonary complications (5), postoperative cardiovascular complications and delirium (25) was observed in elderly patients. In addition, among PD patients with pancreatic cancer, the 60-day mortality of elderly patients aged 75 years or over was greatly increased (26). In a meta-analysis, elderly patients had a high risk of 30-day postoperative mortality and a 50% increased rate of complications following PD (27).
On the contrary, no significant difference in postoperative complications after liver resection was found between elderly and younger patients in a national administrative database in Japan (28). It was also reported that elderly patients had similar perioperative outcomes compared to younger counterparts after PD (29). Unfortunately, the high AAC was not considered in these analyses. A recent study showed that a greater number of components of metabolic syndrome, such as diabetes and dyslipidemia, were significantly associated with AAC (30). We speculated that patients with high AAC levels had a specific background related to an unfavorable lifestyle such as smoking, obesity, diabetes, and hypertension. By evaluating the AAC levels through CT scans, the high-risk patients in major HBP surgery, especially the elderly, would be detected. The AAC level was expected to be an objective index that reflects the patient’s postoperative outcomes in the hospital setting.
We did not analyze long-term complications, such as cardiovascular disease (CVD), in this study. Fortunately, patients within the high AAC group did not develop CVD after short-term follow-up. This study was also limited by its small sample size and single-center design. Larger cohorts, including multi-institutional studies, are necessary to investigate the relationship between AAC and severity of complications after major HBP surgery. With the participation of many cases, it would be possible to analyze the risk factors of severe postoperative complications among high AAC patients for each surgical procedure.
Conclusion
AAC has a stronger correlation with the severity of postoperative complications after major HBP surgery compared to age. This study showed that careful anastomosis and infection control were required to prevent the progression of severe postoperative complications among high-AAC patients. A future prospective study assessing the association between AAC and the decrease in postoperative complications after major HBP surgery is necessary.
Acknowledgements
We thank Editage (www.editage.jp) for the English language review.
Footnotes
Authors’ Contributions
YI and MO conceived and designed the study. YI, MO, KS, KI, TB, and RN acquired the data. YI and MO analyzed and interpreted the data. YI and MO drafted the manuscript. YI, MO, RN, SK, HT, KI, TK, YT, and HO critically revised the article. YI, MO, and HO approved the final version of the manuscript to be published.
Conflicts of Interest
The Authors declare no conflicts of interest.
Funding
This work was supported in part by JSPS KAKENHI Grant Numbers JP20K09104, 22K16534, 22K16535 and AMED under Grant Number 22fk0210108. The funders had no role in the study design, data collection, and analysis. They did not participate in the preparation of the manuscript or the decision to publish.
- Received September 22, 2022.
- Revision received October 4, 2022.
- Accepted October 12, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
