Body composition is associated with postoperative complications in perihilar cholangiocarcinoma

Abstract Background Perihilar cholangiocarcinoma (pCCA) is a malignant tumor of the hepatobiliary system which is still associated with a challenging prognosis. Postoperative complications play a crucial role in determining the overall prognosis of patients with pCCA. Changes in body composition (BC) have been shown to impact the prognosis of various types of tumors. Therefore, our study aimed to investigate the correlation between BC, postoperative complications and oncological outcome in patients with pCCA. Methods All patients with pCCA who underwent curative‐intent surgery for pCCA between 2010 and 2022 were included in this analysis. BC was assessed using preoperative computed tomography and analyzed with the assistance of a 3D Slicer software. Univariate and multivariate binary logistic regression analyses were conducted to examine the relationship between BC and clinical characteristics including various measurements of postoperative complications and Cox regressions and Kaplan–Meier analysis to evaluate oncological risk factors in the study cohort. Results BC was frequently altered in patients undergoing curative‐intent liver resection for pCCA (n = 204) with 52.5% of the patients showing obesity, 55.9% sarcopenia, 21.6% sarcopenic obesity, 48.5% myosteatosis, and 69.1% visceral obesity. In multivariate analysis, severe postoperative complications (Clavien‐Dindo ≥3b) were associated with body mass index (BMI) (Odds ratio (OR) = 2.001, p = 0.024), sarcopenia (OR = 2.145, p = 0.034), and myosteatosis (OR = 2.097, p = 0.017) as independent predictors. Furthermore, sarcopenia was associated with reduced overall survival (OS) in pCCA patients (sarcopenia vs. no‐sarcopenia, 21 months vs. 32 months, p = 0.048 log rank). Conclusions BC is highly associated with severe postoperative complications in patients with pCCA and shows tendency to be associated impaired overall survival. Preoperative assessment of BC and interventions to improve BC might therefore be key to improve outcome in pCCA patients undergoing surgical therapy.


| INTRODUCTION
Cholangiocarcinoma (CCA) is the second most common primary liver cancer after hepatocellular carcinoma (HCC), accounting for 3% of all gastrointestinal cancers. 1 It is classified anatomically into intrahepatic (iCCA), perihilar (pCCA) or distal (dCCA) subtypes with pCCA being the most prevalent form of CCA. 1 Surgery is considered as the primary treatment for pCCA. 2 However, as major liver resection together with complex vascular and biliary reconstruction are commonly necessary in these patient, notable perioperative morbidity, and mortality rates up to 15% have been reported. 3A previously conducted metaanalysis, of 4634 patients, revealed a 40% incidence of severe postoperative complications Clavien-Dindo grade ≥3 after pCCA surgery and representative 90-day mortality of 9%. 4 Also, the cumulative postoperative complications following pCCA resection have a substantial impact on the long-term survival of these patients. 5ancer-associated cachexia is a multifaceted condition characterized by changes in body composition (BC) including weight loss, skeletal muscle wasting (with or without adipose tissue depletion), and a gradual decline in physiological function. 6][9] Postoperative complications have been identified as one of the most important elements affecting postoperative survival in pCCA patients. 5However, only one article had described an association between preoperative skeletal muscle mass exclusively and the incidence of infectious complications in patients with pCCA. 8Alterations in BC, including body mass index (BMI), sarcopenia, myosteatosis, visceral obesity, sarcopenic obesity, and visceral to subcutaneous adipose tissue ratio (VSR) serve as prognostic markers in a variety of clinical settings including the progression of oncological and liver disease. 10][13] Nevertheless, the impact of fluctuations in BC on postoperative complications in patients diagnosed with pCCA remains poorly understood.Until now, only limited research has explored the potential significance of preoperative BC in predicting postoperative complications among patients with pCCA. 8In this study, we conducted a retrospective analysis in a large monocentric cohort to examine the correlation between preoperative BC characteristics and postoperative complications in individuals who underwent surgery for pCCA.We hypothesized that preoperative BC and the occurrence of postoperative complications and long-term outcome are notably associated.

| Study population
We conducted a retrospective study of patients who underwent surgery for pCCA at the RWTH Aachen University Hospital between 2010 and 2022.Exclusion criteria were as follows: (1) patients did not have abdominal CT scans not older than 1 month prior resection, (2) individuals lacking images of the third lumbar vertebral level (L3), (3) patients who relied solely on alternative imaging modalities such as magnetic resonance imaging (MRI).As 10 patients were excluded due to the above-mentioned criteria, the final cohort comprised a total of 204 patients with pCCA.Written informed consent was obtained from the study patients.This study adhered to the principles outlined in the Declaration of Helsinki and received approval from the Ethics Committee of RWTH Aachen University Hospital (EK 23-269).

| Evaluation of computed tomography imaging
A dual-source CT scanner (Siemens Somatom Force, Siemens AG, Munich, Germany) was used for all CT scans.Technical parameters for CT imaging included a tube voltage of 120 kVp, a rotation speed of 0.5 s/rotation and a reconstruction thickness of 5 mm.All CT images underwent contrast enhancement with the images being acquired in the venous phase.Skeletal muscle and adipose tissue in single cross-sectional CT images at the level of the third lumbar vertebra were semi-automatically segmented using the 3D Slicer software platform version 4.1 and the BC module (https:// www.slicer.org/ ) (L3).Attenuation values ranging from −29 to 150 Hounsfield Units (HU) were used to identify and quantify skeletal muscles.Muscle area at the L3 level included various muscles such as the psoas major, erector spinae, lumbar quadrate, transverse abdominis, external and internal obliques, and rectus abdominis.The skeletal muscle index (SMI), calculated by dividing skeletal muscle area (SMA) by the square of height (cm 2 /m 2 ) was used to identify patients with sarcopenia.Visceral fat area (VFA) was assessed using attenuation values between −150 and −50 HU, while subcutaneous adipose tissue was identified using attenuation values between −190 and −30 HU.All measurements were performed by two blinded investigators (GW, JB).In that study, the intra-rater and inter-rater coefficient of variation for SMI was found to be 0.60% and 2.90%, 1.30% and 2.60% for VFA, and 1.20% and 2.40% for SM-RA, respectively.
Historically, a variety of cutoff values have been used for the assessment of BC. 14 The definition of overweight or obesity was based on a BMI ≥25 kg/m 2 .Sarcopenia was diagnosed if the SMI was <41 cm 2 /m 2 in women and < 43 cm 2 /m 2 in men with BMI < 25 kg/m 2 , or if the SMI was <53 cm 2 /m 2 with a BMI ≥25 kg/m 2 .Myosteatosis was diagnosed when SM-RA was <41 HU for BMI < 25 kg/ m2 or < 33 HU for BMI ≥25 kg/m 2 .VFA was considered high if ≥100 cm 2 .Sarcopenic obesity was diagnosed if SMI was ≤38.5 cm2/m2 for women and ≤ 52.4 cm 2 /m 2 for men with BMI ≥25 kg/m 2 .VSR was divided into low VSR (<1.1) and high VSR (≥1.1) (Figure 1).

| Statistical analysis
All statistical analyses were performed using SPSS software (IBM SPSS Statistics, version 26.0).Data are presented as numbers and percentages or median and interquartile ranges as data distribution might not be distributed normally in every subanalysis.Univariate and multivariate logistic regression analyses were performed to investigate the relationships between different clinical parameters in BC.Major complications were defined as any complications rated ≥3b according to the Clavien-Dindo Scale.The Kaplan-Meier method was used to estimate overall survival (OS) and recurrence-free survival (RFS).Cox regressions were performed to identify the prognostic factors associated with OS and RFS.Statistical significance was set at p < 0.05.

| Patient characteristics
A total of 204 patients diagnosed with pCCA were enrolled in this study.The median age of the population was 68 years with most of the individuals being sassed as ASA III or higher (127/204, 62.3%).A variety of operative procedures with median operation time of 431 min was conducted.R0 resection was achieved in 164 (80.4%) of the patients.][17][18] Regarding, overall complications, 120 (58.8%) had a displayed complications Clavien-Dindo grade ≤ 3A and 84 (41.2%) grade ≥ 3B.The perioperative mortality rate was 15.7% (32/206).More details regarding the study cohort are displayed Table 1.

| Associations of body composition to clinical characteristics and complications
In the study cohort, 107 patients (52.5%) presented with obesity, 114 patients (55.9%) were classified as sarcopenic, 99 patients (48.5%) exhibited myosteatosis, 141 patients (69.1%) displayed visceral obesity, 44 patients (21.6%) demonstrated sarcopenic obesity and 66 patients (32.4%) presented with a high VSR (Table 1).To assess the relationship between these BC characteristics and clinical features including postoperative complications, univariate and multivariate logistic regression analyses were conducted.To ensure the validity of results all above mentioned measures for complications were included in the analysis among other known clinical variables.
The detailed results of the analysis are displayed in Table 2 and Table S1.The brief results of the multivariate analysis were as followed: Individuals with obesity showed low levels of alkaline phosphatase (odds ratio, (OR) = 0.081, p = 0.019), reduced lymph node involvement (OR = 0.507, p = 0.025) and an increased likelihood of experiencing severe postoperative complications (OR = 2.001, p = 0.024).Patients with sarcopenia were more likely to be female (OR = 8.102, p<0.001) and did also experience more serious post-operative complications (OR = 2.145, p = 0.034).
Patients were further stratified for survival analysis based on the concurrent presence or absence of sarcopenia and myosteatosis.Compared to other patients, those with both conditions had a significantly shorter OS (27 months vs. 39 months, p = 0.017, log-rank test).However, there was no statistically significant difference in RFS between the rest of the cohort and patients with both conditions (37 months vs. 35 months, p = 0.916, logrank test) (Figure 4).

| Survival analysis by Cox regressions
To investigate the relationship between BC and clinical parameters with oncological outcomes, univariate and multivariate Cox regression analyses were conducted.
In this study, we aimed to assess the association between preoperative BC and postoperative complications in pCCA patients undergoing curative-intent surgery.Severe malnutrition and cachexia, commonly observed in individuals with cancer, detrimentally impact the prognosis and can increases the likelihood of infections and mortality in this patient population. 19Additionally, it may contribute to a higher occurrence of cancer-related complications. 20espite notable advancements in intraoperative and perioperative care in recent years, postoperative complications continue to significantly impact patients short-and longterm quality of life following surgery. 5Thus, effectively reducing the occurrence of postoperative complications and enhancing the curative outcomes for pCCA have emerged as an area of significant clinical concern.
In our study, we observed a notable proportion of patients diagnosed with pCCA to exhibit considerable alterations in BC.Specifically, 52.5% of the study set displayed obesity, 55.9% sarcopenia, 48.5% myosteatosis, 69.1% visceral obesity, 21.6% sarcopenic obesity and 32.4% high VSR.Regarding postoperative complications, a substantial association was shown between obesity, sarcopenia, myosteatosis and the actual occurrence of severe postoperative complications (Clavien-Dindo≥3b).As severe postoperative complications were also associated with impaired OS, the clinical importance of altered BC is underlined by our data.Furthermore, Kaplan-Meier analysis showed a correlation between sarcopenia and OS in pCCA patients.Crucially, our findings illuminate the potential influence of BC on both postoperative complications and long-term survival in pCCA, providing a novel contribution to the field.This study does further constitute the first comprehensive analysis of various measures of BC in the context of complications after surgery for pCCA and the first evidence that myosteatosis is a major predictor in postoperative outcome in these patients.
The occurrence of postoperative complications significantly impacts the prognosis following CCA surgery. 21herefore, mitigating risk factors associated with these complications might help to enhance the prognosis of patients with CCA.Recent studies have identified several risk factors linked to postoperative complications in this patient population.Yuichi et al. demonstrated that patients with low psoas muscle index (PMI) or psoas muscle density (PMD) were more prone to infectious complications after surgery, particularly among those with pCCA. 8umamoto et al. discovered that BMI and neutrophil to lymphocyte ratio (NLR) were significant predictors of postoperative complications. 22Other reports have indicated that preoperative 6-min walk distance are associated with postoperative complications in individuals with CAA. 23

Multivariate analysis
Liver failure (No/Yes(%);ref  A comprehensive retrospective study indicates a potential correlation between obesity and both surgical wound infections and extended operative duration. 24Our findings validate the notion that significant postoperative complications in pCCA patients are linked to BMI ≥25 kg/ m 2 , which aligns with previous research demonstrating a comparable impact of high BMI on postoperative complications in distal CCA treated by pancreatic resections. 22is is also consistent with our previous observations and findings. 10Impaired immune function may also contribute to the risk of severe complications in obese individuals. 25urthermore, obesity is associated with reduced therapeutic tissue levels of applied antibiotics which can play a role in the effective management of complications. 26arcopenia, characterized by progressive muscle mass and function loss, is commonly observed in the elderly F I G U R E 2 Recurrence-free in relation to body composition characteristics in patients with perihilar cholangiocarcinoma.RFS for (A) BMI (BMI<25 kg/m 2 vs. BMI≥25 kg/m 2 : 52 vs. 25 months), (B) sarcopenia (no-sarcopenia vs sarcopenia: 29 vs. 39 months), (C) myosteatosis (no-myosteatosis vs. myosteatosis: 31 vs. 36 months), (D) visceral obesity (no-visceral obesity vs. visceral obesity: 36 vs. 35 months) and (E) sarcopenic obesity (no-sarcopenic obesity vs. sarcopenic obesity: 31 vs. 42 months), (F) VSR (low VSR vs. high VSR: 37 vs. 29 months).BMI, body mass index; OS, over survival; RFS, recurrence free survival; VSR, visceral to subcutaneous adipose tissue ratio.patients with malignancies. 27It can result from an imbalance between insufficient food intake and high tumor metabolism, leading to a negative energy balance. 27Patients with pCCA are particularly susceptible to obstructive jaundice, which can also contribute to inadequate intake and reduced physical activity, thereby inducing sarcopenia. 28arcopenia has been identified as a significant risk factor for poor perioperative outcomes undergoing hepatectomy within 90 days in previous studies, further supporting our findings as we have observed a strong association between sarcopenia and severe postoperative complications. 29nterestingly, previous studies have also shown that sarcopenia does not appear to be associated with postoperative complications in patients with pCCA, likely due to the limited sample size of the previous studies. 7,10yosteatosis is a pathological condition characterized by the accumulation of fat within muscle tissue, leading to impaired muscle and mitochondrial function. 30A F I G U R E 3 Overall survival in relation to body composition characteristics in patients with perihilar cholangiocarcinoma.OS for (A) BMI (BMI<25 kg/m 2 vs. BMI≥25 kg/m 2 : 31 vs. 25 months), (B) sarcopenia (no-sarcopenia vs sarcopenia: 31 vs. 21 months), (C) myosteatosis (no-myosteatosis vs. myosteatosis: 25 vs. 27 months), (D) visceral obesity (no-visceral obesity vs. visceral obesity: 25 vs. 27 months) and (E) sarcopenic obesity (no-sarcopenic obesity vs. sarcopenic obesity: 25 vs. 28 months), (F) VSR (low VSR vs. high VSR:39 vs. 31 months).BMI, body mass index; OS, over survival; RFS, recurrence free survival; VSR, visceral to subcutaneous adipose tissue ratio.meta-analysis has revealed that myosteatosis as a significant predictor of survival in various types of cancer, with a 73% higher mortality rate reported in cancer patients with myosteatosis compared to those without. 31Our study provides novel evidence suggesting that myosteatosis is also an important predictor of serious complications following surgery for pCCA.This association may be explained by the inhibitory effects of myosteatosis on glutamine synthetase and its role in increasing inflammatory cytokines, ultimately impairing ammonia detoxification, a critical process for maintaining physiological balance. 32ymph node metastasis represents the predominant form of metastasis in pCCA, thus making lymph node dissection a crucial component of pCCA surgery.Notably, lymph node metastasis in patients with pCCA significantly influences rates of early postoperative recurrence. 33Patients without lymph node metastasis exhibit a superior overall survival compared to those with positive lymph nodes. 34Moreover, nodal status correlates with disease-specific survival. 35Interestingly, our study indicates a higher likelihood of lymph node metastasis in obese patients, while patients with sarcopenic obesity show a reduced likelihood.Further exploration is required to understand the underlying reasons for this observation.Additionally, it is worth noting that the median survival in patients with lymph node metastases does not differ between those who underwent complete (R0) or incomplete (R1) resection. 36n patients diagnosed with pCCA, liver function is commonly compromised due to preoperative cholestasis and cholangitis.In a study conducted by Ribero et al., which included a cohort of 133 pCCA patients, the occurrence of preoperative cholangitis was identified as a significant risk factor for postoperative hepatic failure and mortality following hepatectomy. 37Subsequent studies have reported similar findings, linking preoperative cholangitis in pCCA patients to increased postoperative mortality. 38However, the precise underlying pathophysiological mechanisms remain unknown.In our study, an association between the presence of preoperative cholangitis and myosteatosis was observed.Based on these findings, we hypothesize that preoperative cholangitis might be one of the factors influencing the development of myosteatosis and thereby, subsequently impacting the prognosis of patients with pCCA.
Furthermore, sarcopenia was found to have a negative impact on OS, but no significant correlation with RFS was observed, consistent with earlier research. 29oth sarcopenia and myosteatosis represent a loss of muscle function and quality.Inflammatory responses and oxidative stress are primary etiological factors in muscle wasting. 39Inflammatory cytokines can disrupt the balance between muscle protein synthesis and degradation, leading to muscle wasting, 40 including chronic inflammation driven by interleukin-1, interleukin-6, and tumor necrosis factor-alpha, which are associated with sarcopenia. 41Sarcopenia as one of the features of poor prognosis in cancer may be attributed to systemic inflammation and cachexia induced by the malignancy.An intriguing role for muscle tissue is to modulate immune regulation through the secretion of peptides and cytokines. 42The maintenance of immune function is further supported by muscle production of the cytokines interleukin-7 (IL-7) and interleukin-15 (IL-15). 43onsequently, we postulate that the systemic inflammatory response and alterations in immune function triggered by myosteatosis and sarcopenia are associated with postoperative complications and OS in patients with pCCA.In a subgroup analysis, we also found that patients with concurrent myosteatosis and sarcopenia F I G U R E 4 Overall and recurrence-free survival with respect to the combination of sarcopenia and myosteatosis.(A) Recurrence-free survival in patients with sarcopenia and myosteatosis (median RFS, no sarcopenia or no myosteatosis vs. sarcopenia and myosteatosis: 37 vs. 35 months).(B) Overall survival in patients sarcopenia and myosteatosis (median OS, no sarcopenia or no myosteatosis vs. sarcopenia and myosteatosis: 39 vs. 27 months).OS, overall survival.RFS, recurrence-free survival.similarly affect OS.We hypothesize that the concurrent presence of myosteatosis and sarcopenia may exert a compounded effect on OS.
Our study has shown that BC significantly influences the occurrence of postoperative complications in pCCA, which in turn impacts the prognosis of patients with pCCA.Previous research has underscored the vital role of muscle health in improving the prognosis of surgical patients, emphasizing the need for a comprehensive assessment of patients' nutritional status and muscle composition to enhance surgical outcomes. 44Therefore, improving BC becomes also crucial for pCCA patients.As such, Raff et al. discussed a prehabilitation program designed for cancer patients undergoing neoadjuvant treatment, aimed at improving patients' physical, nutritional, and psychological status before surgery. 45Some studies suggest that a combination of physical activity and nutritional supplementation promotion may be an effective intervention for patients with sarcopenia, which is often associated with pCCA. 46The engagement in physical exercise also contributes to the enhancement of skeletal muscle strength and quality and attenuate inflammatory responses, potentially preventing complications. 47,48Perioperative nutritional supplementation has the potential to reduce the risk of postoperative complications and shorten the length of hospital stay for patients. 49In a study involving patients undergoing surgery for pancreatic cancer, physical functioning showed improvement by a home-based rehabilitation program implemented both before and after the surgery. 50eucine is known to directly initiate the translation of mRNA and plays a critical role in the regulation of muscle protein synthesis in both humans and rodents. 51iven the pivotal role of leucine in protein synthesis, supplementation with this amino acid may be beneficial in enhancing muscle protein synthesis.Previous studies have shown that a short-term preoperative exercise program combined with the consumption of a leucineenriched beverage may contribute to increases in body weight, skeletal muscle mass, and SMI, thereby improving patient prognosis. 52However, there is currently a lack of research on strategies to improve BC specifically in patients with CCA.Further experiments and studies are needed to address this gap and analyze potential interventions in the future.
There are several limitations to this study.It is important to note that this was a retrospective study conducted at a single hepatobiliary center.Although the sample size is sufficient, further validation in multiple institutions or countries is necessary to enhance the generalizability of the findings.Further, the impact of rehabilitation on the preoperative or postoperative improvement of BC parameters in patients with pCCA remains unclear.Thus, the relationship between these improvements and their potential effects on reducing postoperative complications and long-term survival needs further investigation.Addressing these clinical questions will require multiinstitutional randomized controlled studies in the future.Despite our efforts to control for multiple potential confounding factors by multivariable analysis, as with all observational studies, there may still be variables that are either unidentified or not adequately adjusted.Further, it must be noted that we exclusively addressed patients who underwent surgical resection for pCCA.Thus, our data cannot be used to draw conclusions for the wider population of pCCA patients who are not surgical candidates due to local irresectability or distant metastases.
Future studies should also emphasize the phenomenon of changes of BC over time both preoperatively as well as in the postoperative setting during oncological follow-up.Changes of BC over time might even give more information about the individual patient's prognosis and susceptibility to postoperative complications.Unfortunately, our retrospective data set does not allow a longitudinal assessment in the study patients.This should be addressed in future prospective clinical cohorts.

| CONCLUSION
Our study indicates that changes in BC are strongly linked to postoperative complications in pCCA patients.Consequently, assessing BC prior to surgery in pCCA patients could serve as a valuable tool for risk perioperative stratification.In the future, it is essential to focus on monitoring BC changes before surgery in pCCA cases and develop suitable nutritional and exercise programs to enhance BC prior to the actual surgical procedure.By addressing BC status preoperatively, we may potentially optimize patient outcomes and minimize the risk of postoperative complications.To achieve this goal, further research and the development of tailored interventions are needed to effectively improve BC in pCCA patients.
Note: Data presented as median and interquartile range if not noted otherwise.Abbreviations: ALPPS, associating liver partition with portal vein ligation for staged hepatectomy; ALT, alanine aminotransferase; ASA, American Society of Anesthesiology; AST, aspartate aminotransferase; AP, alkaline phosphatase; BMI, body mass index; CRP, C-reactive protein; F, female; FFP, fresh frozen plasma; GGT, gamma-glutamyl transferase; INR, international normalized ratio; LVI, lymph vascular invasion; M, male; MVI, microvascular invasion; OS, overall survival; PRBC, packed red blood cells; RFS, recurrence-free survival; SMI, skeletal muscle index; SM-RA, skeletal muscle radiation attenuation; VSR, visceral to subcutaneous adipose tissue ratio.T A B L E 1 (Continued) | 7 of 18 WANG et al.T A B L E 2

TABLE 2 (
Continued) Analysis of recurrence-free survival and overall survival in perihilar cholangiocarcinoma.