Access to this full-text is provided by Frontiers.
Content available from Frontiers in Surgery
This content is subject to copyright.
EDITED BY
Zoltan Czigany,
Charité Universitätsmedizin Berlin, Germany
REVIEWED BY
Tevfiktolga Sahin,
I
nönü University, Turkey
Katharina Jöchle,
University Hospital RWTH Aachen, Germany
Decan Jiang,
Charité Universitätsmedizin Berlin, Germany
*CORRESPONDENCE
Wenbo Meng
mengwb@lzu.edu.cn
Ping Yue
dryueping@sina.com
†
These authors have contributed equally to this
work
SPECIALTY SECTION
This article was submitted to Surgical
Oncology, a section of the journal Frontiers in
Surgery
RECEIVED 07 November 2022
ACCEPTED 02 December 2022
PUBLISHED 10 January 2023
CITATION
He Y, Liu H, Ma Y, Li J, Zhang J, Ren Y, Dong C,
Bai B, Zhang Y, Lin Y, Yue P and Meng W (2023)
Preoperative prognostic nutritional index
predicts short-term complications after radical
resection of distal cholangiocarcinoma.
Front. Surg. 9:1091534.
doi: 10.3389/fsurg.2022.1091534
COPYRIGHT
© 2023 He, Liu, Ma, Li, Zhang, Ren, Dong, Bai,
Zhang, Lin, Yue and Meng. This is an open-
access article distributed under the terms of the
Creative Commons Attribution License (CC BY).
The use, distribution or reproduction in other
forums is permitted, provided the original
author(s) and the copyright owner(s) are
credited and that the original publication in this
journal is cited, in accordance with accepted
academic practice. No use, distribution or
reproduction is permitted which does not
comply with these terms.
Preoperative prognostic
nutritional index predicts
short-term complications
after radical resection of distal
cholangiocarcinoma
Yulong He1,2†, Haoran Liu1,2†, Yuhu Ma1,2, Jianlong Li1,2,
Jinduo Zhang1,2, Yanxian Ren1,2, Chunlu Dong1,2, Bing Bai1,2,
Yong Zhang2, Yanyan Lin1,2, Ping Yue1,2,3*and Wenbo Meng1,2,3*
1
The First Clinical Medical College, Lanzhou University, Lanzhou, China,
2
Department of General
Surgery, The First Hospital of Lanzhou University, Lanzhou, China,
3
Gansu Province Key Laboratory of
Biological Therapy and Regenerative Medicine Transformation, Lanzhou, China
Background: The occurrence of postoperative complications of distal
cholangiocarcinoma (dCCA) is an indicator of poor patient prognosis. This
study aimed to determine the immune-nutritional indexes (INIs) that can
predict short-term postoperative complications.
Methods: A retrospective analysis of 148 patients with dCCA who were
operated radical pancreaticoduodenectomy at the First Hospital of Lanzhou
University from December 2015 to March 2022 was conducted to assess the
predictive value of preoperative INIs and preoperative laboratory tests for
short-term postoperative complications, and a decision tree model was
developed using classification and regression tree (CART) analysis to identify
subgroups at risk for overall complications.
Results: In this study, 83 patients (56.08%) experienced overall complications.
Clavien-Dindo grade III-V complications occurred in 20 patients (13.51%), and
2 patients died. The areas under curves (AUCs) of the preoperative prognostic
nutritional index (PNI), controlling nutritional status (CONUT) score, and
neutrophil-to-lymphocyte ratio (NLR) were compared; the PNI provided the
maximum discrimination for complications (AUC = 0.685, 95% CI = 0.600–
0.770), with an optimal cutoff value of 46.9, and the PNI ≤46.9 group had
higher incidences of overall complications (70.67% vs. 40.00%, P<0.001) and
infectious complications (28.77% vs. 13.33%, P= 0.035). Multivariate logistic
regression analysis identified PNI (OR = 0.87, 95% CI: 0.80–0.94) and total
bilirubin (OR = 1.01, 95% CI: 1.00–1.01) were independent risk factors for
overall complications (P< 0.05). According to CART analysis, PNI was the most
important parameter, followed by the total bilirubin (TBIL) level. Patients with a
PNI lower than the critical value and TBIL higher than the critical value had
the highest overall complication rate (90.24%); the risk prediction model had
an AUC of 0.714 (95% CI, 0.640–0.789) and could be used to stratify the risk
of overall complications and predict grade I-II complications (P<0.05).
Abbreviations
CCA, cholangiocarcinoma; dCCA, distal cholangiocarcinoma; INIs, immune-nutritional indexes; PNI,
prognostic nutritional index; CONUT, nutritional control status; NLR, neutrophil-to-lymphocyte ratio;
CART, Classification And Regression Tree; TBIL, total bilirubin; ROC, receiver operating characteristic;
AUC, area under the curve.
TYPE Original Research
PUBLISHED 10 January 2023
|
DOI 10.3389/fsurg.2022.1091534
Frontiers in Surgery 01 frontiersin.org
Conclusion: The preoperative PNI is a good predictor for short-term complications after
the radical resection of dCCA. The decision tree model makes PNI and TBIL easier to use
in clinical practice.
KEYWORDS
distal cholangiocarcinoma, postoperative complication, prognostic nutritional index, total
bilirubin, decision tree
Introduction
Cholangiocarcinoma (CCA) is a highly lethal malignancy
that may occur anywhere within the biliary tree and/or liver
parenchyma (1). The incidence of CCA is gradually increasing
worldwide (2). Depending on the anatomical site of origin,
CCA is classified as intrahepatic, perihilar, or distal
cholangiocarcinoma (dCCA), and dCCA accounts for
approximately 20% of CCA cases (3). Due to the location and
invasive nature of dCCA, such patients often present with
locally advanced or metastatic disease. Surgery through
pancreaticoduodenectomy is the only curative option (4).
Postoperative complications are one of the factors for poor
prognosis in many cancers, and the occurrence of
postoperative complications will prolong the postoperative
recovery time, increase the economic burden, and result in
poor long-term prognosis (5). It has been reported that
postoperative complications may lead to systemic
inflammation, which may reduce the immune response to
cancer (6).
Patients with cancer often suffer from severe nutritional
deficiencies, malnutrition, and low immunity will adversely affect
the development of cancer patients, increase the incidence of
infection, length of hospital stay, and risk of death (7), and may
lead to higher rates of complications (8). In addition, We
reviewed the relationship between preoperative immuno-
nutritional related indicators and the occurrence of
complications and prognosis in patients with CCA, where the
association of prognostic nutritional index (PNI), nutritional
control status (CONUT) score, neutrophil-to-lymphocyte ratio
(NLR)and dCCA needs to be further investigated. This research
aimed to evaluate the ability of preoperative immune-nutritional
indexes (INIs) to predict postoperative complications after
radical resection of dCCA and establish a decision tree model
subgroups according to risk level to provide a basis for the
prevention and treatment of postoperative complications, guide
health education and prognosis management for patients, and
accelerate patient recovery.
Methods
This retrospective study wasconducted in The First Hospital of
Lanzhou University in China, the studyfollowed the Declaration of
Helsinki and was approved by the Ethics Committee of The First
Hospital of Lanzhou University (LDYY2022-412).
Patient enrollment
This retrospective study included patients with
pancreaticoduodenectomy for dCCA at the First Hospital of
Lanzhou University from December 2015 to March 2022.
Patients who met the following criteria were included: (1)
postoperative pathological examination consistent with a
diagnosis of cholangiocarcinoma; (2) no distant metastasis
found during the operation; (3) no severe heart, lung,
kidney, or brain dysfunction before the operation (without
combined organ dysfunction such as heart failure,
respiratory failure, renal failure, disorientation and stress
disorder, etc.); and (4) complete clinical medical records.
Subjects with the following were excluded: (1) combined
vascular resection and reconstruction; (2) extended radical
resection combined with resection of other organs; (3)
postoperative pathology showing a positive cut margin; (4)
previously complicated with other malignant tumors; and
(5) lack of clinical medical records. Finally, 148 patients
were enrolled in this study (Figure 1).
Data collection
Hospital records were retrospectively evaluated for
baseline information, laboratory tests, postoperative
pathological factors, and postoperative complications, as
well as PNI, CONUT score, and NLR when the patient was
first admitted to the hospital. PNI was calculated using the
following formula: serum albumin (g/L) + 5 × total
lymphocyte count (10
9
L
−1
)(9).TheCONUTscoreconsists
of three parameters: serum albumin, total peripheral
lymphocyte count, and total cholesterol. Postoperative
complications were defined as those occurring within 2
weeks after surgery and were assessed by the Clavien‒
Dindo classification (10). Abdominal infection was defined
by a positive abdominal culture collection, persistent fever
or elevated leucocyte count, and effective antibiotic
treatment, with or without bacteremia or a liver abscess,
together with the exclusion of other infections including
He et al. 10.3389/fsurg.2022.1091534
Frontiers in Surgery 02 frontiersin.org
pulmonary and wound infections (11). Postoperatively,
delayed gastric emptying was defined as the need for gastric
tube placement for more than 3 days after the operation,
the need for repeat catheter placement due to vomiting and
other reasons after extubation, and an inability to consume
solids for 7 days or longer after the operation (12).
Statistical analysis
All statistical analyses in this study were conducted using
R (version 4.1.3). Quantitative data are presented as median
(interquartile range) and were compared using the Wilcoxon
rank sum test. Qualitative data were compared by the chi-square
test and are expressed as frequency and percentage (%).
Pvalues < 0.05 were considered statistically significant. The
potential predictive value of preoperative INIs for overall
complications was assessed by the area under the receiver
operating characteristic (ROC) curve. The Youden index was
used to select the optimal cutoff value, which was set as the
value with the maximum summed sensitivity and specificity
value. Logistic regression analysis was used to identify risk
factors for overall complications. Pvalues<0.05wereconsidered
statistically significant, and independent risk factors for overall
complications after surgery for dCCA were identified by
univariate and multifactorial analysis.
The classification and regression tree (CART) model was
constructed using the R package “party”(version 4.1.3) by
identifying independent risk factors for overall complications
through logistic regression analysis. CART, a machine
learning method for constructing predictive models that
simulate clinical decision-making processes (13), is easy to
understand and implement. The decision tree construction
process is recursive. Thus, the parameter most closely related
FIGURE 1
Patient enrollment flowchart.
He et al. 10.3389/fsurg.2022.1091534
Frontiers in Surgery 03 frontiersin.org
to the result is extracted as the first node, When there is no
correlation between the predictor variable and the outcome
variable, the algorithm stops, and the pvalues show the
relationship between the preoperative variables and the
postoperative complications. Finally, the AUC was used to
evaluate the accuracy of the decision tree in predicting overall
complication outcomes.
Results
Patient characteristics
This study included 148 patients, 88 males, and 60 females,
the age range of 28–78 years. Clinical characteristics, such as
preoperative blood routine, biochemical tests, coagulation
TABLE 1 Clinical characteristics of the patients at baseline.
Characteristics Complication (n= 83) N-Complication (n= 65) PValue
Age (years), median (IQR) 63 (57.5–61) 61 (54–69) 0.234
Male sex, n(%) 53 (63.86%) 35 (53.85%) 0.288
Laboratory examination, median (IQR)
Erythrocyte (109/L) 4.31 (4.04–4.71) 4.41 (4.02–4.86) 0.354
Leukocyte (109/L) 6.35 (5.08–9) 5.30 (4.53–6.64) 0.002
Hemoglobin (g/L) 137 (131–148.5) 139 (126–154) 0.857
Albumin (g/L) 39.7 (36.8–42.9) 41 (39–44) 0.006
Total lymphocyte (109/L) 1.27 (0.97–1.41) 1.4 (1.22–1.7) 0.006
Aspartate transaminase (U/L) 129 (77–195.5) 114 (74–251) 0.845
Total bilirubin (µmol/L) 267.9 (162.35–368.1) 154.8 (90–219) <0.001
Cholesterol (mmol/L) 4.77 (4.12–6.47) 4.87 (3.94–6.15) 0.759
Triglycerides (mmol/L) 1.71 (1.12–2.47) 2.11 (1.56–2.94) 0.030
Prothrombin time (S) 11.4 (10.8–12.8) 11.2 (10.7–12.1) 0.050
Carbohydrate antigen 199 (U/ml) 129 (35.75–272.55) 86.9 (56.5–190) 0.655
Comorbidities, n(%)
Diabetes 6 (7.23%) 6 (7.23%) 0.889
Hypertension 18 (21.69%) 16 (24.62%) 0.823
Bile duct stones 8 (9.64%) 3 (4.62%) 0.401
Cholecystolithiasis 10 (12.05%) 7 (10.78%) 1.000
Hepatitis 4 (4.82%) 3 (4.62%) 1.000
Pathological factors, n(%)
TNM staging 0.905
0-IIb 62 (74.70%) 50 (76.92%)
IIb-IV 21 (25.30%) 15 (23.08%)
Vascular invasion 17 (20.48%) 10 (15.38%) 0.589
Lymph node metastasis 28 (33.73%) 18 (27.69%) 0.542
Differentiation 0.008
Well 7 (8.43%) 5 (7.69%)
Moderate 23 (27.72%) 14 (21.54%)
Low-middle 48 (57.83%) 29 (44.62%)
Poorly 5 (6.02%) 17 (26.98%)
Data are presented as n(%) or median (IQR). IQR, Interquartile range.
He et al. 10.3389/fsurg.2022.1091534
Frontiers in Surgery 04 frontiersin.org
indicators, tumor markers, and postoperative pathology, were
collected, and the clinical characteristics of the patients were
summarized and analyzed (Table 1).
In this study, 83 patients (56.08%) had overall complications, of
which 20 patients (13.51%) had Clavien‒Dindo grade III-V
complications, and 2 patients died. There were 60 cases (40.54%)
of pancreatic leakage, 35 cases (23.65%) of biochemical leakage, 24
cases (16.22%) of Grade B pancreatic leakage, 1 case (0.68%) of
Grade C pancreatic leakage, 33 cases (22.30%) of delayed gastric
emptying, 23 cases (15.54%) of abdominal infection, and 18 cases
(12.26%) of abdominal bleeding. There were 13 cases (8.78%) of
bile leakage, 11 cases (7.43%) of pulmonary infection, and 4 cases
(2.70%) of wound infection (Table 2).
Predictive values of INIs for postoperative
complications
For investigating the predictive value of preoperative INIs for
short-term complications after radical surgery for dCCA and
evaluating the correlation between the predictor and outcome
variables, we plotted ROC curves for PNI, CONUT, and NLR.
The results show that PNI has the highest differentiation (AUC
= 0.685, 95% CI = 0.600–0.770), followed by NLR (AUC = 0.678,
95% CI = 0.592–0.765) and CONUT (AUC = 0.603, 95% CI =
0.514–0.692). A comparison of the AUC of each curve showed
that PNI correlated more strongly with complications than
CONUT and NLR. PNI has a better predictive value for
postoperative complications, with an optimal cutoff value of
46.9, corresponding to the maximum Youden index (Figure 2).
Correlations between PNI and clinical
characteristics
The patients were divided into a low PNI group (PNI ≤46.9,
n= 73) and a high PNI group (PNI > 46.9, n= 75) based on the
optimal cutoff value. The low PNI group had lower erythrocyte
counts, higher leukocyte counts, lower hemoglobin levels, lower
albumin levels, lower total lymphocyte counts, higher total
bilirubin levels, and longer prothrombin times. There were no
significant differences in pathological characteristics between the
two groups. In terms of postoperative complications, the low
PNI group had a higher occurrence of complications (70.67%,
P< 0.001), and a low PNI had a stronger ability to predict
TABLE 2 Postoperative complications.
All (n= 148) CD Grades I-II CD Grades III-V
Postoperative complication 83 (56.08%) 63 (42.57%) 20 (13.51%)
Intraperitoneal hemorrhage 18 (12.16%) 4 (2.70%) 14 (9.46%)
Pancreatic leakage 60 (40.54%) 46 (31.08%) 14 (9.46%)
Biochemical leakage 35 (23.65%) 31 (20.95%) 4 (2.70%)
Grade B 24 (16.22%) 15 (10.14%) 9 (6.08%)
Grade C 1 (0.68%) 0 1 (0.68%)
Bile leakage 13 (8.78%) 8 (5.41%) 5 (3.37%)
Abdominal infection 23 (15.54%) 12 (8.11%) 11 (7.43%)
Wound infection 4 (2.70%) 2 (1.35%) 2 (1.35%)
Pulmonary infection 11 (7.43%) 4 (2.70%) 7 (4.73%)
Delayed gastric emptying 33 (22.30%) 25 (16.89%) 8 (5.41%)
Data are presented as n(%). CD, Clavien–Dindo classification.
FIGURE 2
ROC curves for INIs related to postoperative complications.
He et al. 10.3389/fsurg.2022.1091534
Frontiers in Surgery 05 frontiersin.org
overall complications [Figure 3(a)], and the low PNI group had a
higher occurrence of infectious complications (28.77%, P= 0.035)
[Figure 3(b)]. A low PNI had predictive value for overall
complications and infectious complications (Table 3).
Logistic regression analysis to identify risk
factors for overall complications
This study used logistic regression analysis to identify risk
factors for postoperative complications (Table 4). Univariate
analysis showed that prognostic nutritional index (OR = 0.85,
95% CI: 0.79–0.93), total bilirubin (OR = 1.00, 95% CI:
1–1.01), leukocyte count (OR = 1.28, 95% CI: 1.09–1.49),
prothrombin time (OR = 1.34, 95% CI: 1.06–1.70) and
triglycerides (OR = 1.51, 95% CI: 1.04–2.19) were influencing
factors for overall complications (P< 0.05). The above
influencing factors were included in multivariate logistic
regression analysis, and the results showed that prognostic
nutritional index (OR = 0.87, 95% CI: 0.80–0.94) and total
bilirubin (OR = 1.01, 95% CI: 1.00–1.01) were independent
risk factors for overall complications (P< 0.05).
CART-based subgroup analysis of overall
complication risk
In the CART algorithm, we used the important variables
PNI and total bilirubin (TBIL). The PNI was the most critical
parameter for overall complications, and the optimal cutoff
value was 46.9. Based on the correlation between the predictor
and outcome variables, the CART algorithm selected the
optimal cut-off point for TBIL levels for all patients, with an
optimal cut-off value of 194.1 µmol/L (Figure 4). Therefore,
patients with a PNI lower than the cutoff value and TBIL
higher than the cutoff value had the highest complication rate
(high-risk group: 37/41 patients, 90.24%), followed by the
medium-risk group and low-risk group having overall
complication rates of 50% (16/32 patients) and 40% (30/75
patients), respectively. The risk prediction model had an AUC
of 0.714 (95% CI, 0.640–0.789) and had a good effect in
predicting the overall risk of complications (Figure 5). We
analyzed the relationship between risk groups and Clavien‒
Dindo grades established by the decision tree model
(Table 5), which could stratify overall complication risk
groups and predict grade I-II complications (P< 0.05).
Discussion
The prognosis of dCCA is poor, and the median survival time of
patients with the unresectable disease is less than one year (14).
Patients with dCCA who underwent radical
pancreaticoduodenectomy, especially R0 resection, can achieve
long-term survival (15). However, postoperative complications
may lead to the deterioration of long-term survival, more use of
hospital resources, and an increase in the postoperative
FIGURE 3
(A) Frequency of postoperative complications; (B) Frequency of infectious complications.
He et al. 10.3389/fsurg.2022.1091534
Frontiers in Surgery 06 frontiersin.org
readmission rate (16). Effective preoperative differentiation and
intervention of high-risk patients can be of great benefittoboth
clinicians and patients, and this study focuses on identifying
preoperative PNI and TBIL levels as valid indicators for
predicting short-term complications after radical surgery for dCCA.
The PNI was first proposed in 1980 by Onodera et al. (17)
for preoperative nutritional status assessments and surgical risk
predictions in gastrointestinal surgery. Serum albumin has been
recognized as a prognostic indicator of various diseases and has
been widely used to assess nutritional status and prognosis
(18,19). Lymphocytes play an important role in tumor-related
immunology and have strong antitumor immune functions
(20). Therefore, PNI reflects the nutritional and immune
status of patients and has been widely used as a prognostic
TABLE 3 Clinicopathological characteristics of the low PNI group and the high PNI group.
Characteristics PNI ≤46.9 (n= 73) PNI > 46.9 (n= 75) PValue
Laboratory examination, median (IQR)
Erythrocyte (10
9
/L) 4.22 (3.93–4.55) 4.6 (4.09–5.02) 0.015
Leukocyte (10
9
/L) 6.35 (5.1–9.13) 5.44 (4.75–6.99) 0.023
Hemoglobin (g/L) 137 (122–145) 141 (130–145) 0.022
Albumin (g/L) 38.8 (35.9–39.9) 43.3 (41.05–45.3) <0.001
Total lymphocyte (10
9
/L) 1.27 (0.87–1.37) 1.42 (1.22–1.87) <0.001
Aspartate transaminase (U/L) 146 (75–260) 118 (74–198.5) 0.444
Total bilirubin (µmol/L) 223.2 (137.5–380.5) 194.1 (124.65–262.9) 0.049
Cholesterol (mmol/L) 4.87 (4–6.15) 4.79 (4.03–6.49) 0.766
Triglycerides (mmol/L) 1.72 (1.34–2.76) 2.04 (1.27–2.85) 0.914
Prothrombin time (S) 11.5 (11–12.9) 11 (10.55–12.15) 0.002
Carbohydrate antigen 199 (U/ml) 136.1 (50.3–283.4) 82.1 (43.05–183.45) 0.163
Pathological factors, n(%)
TNM staging 0.922
0-IIb 56 (76.71%) 56 (74.67%)
IIb-IV 17 (23.29%) 19 (25.33%)
Vascular invasion 10 (13.70%) 17 (22.67%) 0.230
Lymph node metastasis 17 (23.29%) 29 (38.67%) 0.065
Differentiation 0.889
Well 7 (9.59%) 5 (6.67%)
Moderate 19 (26.03%) 18 (24.00%)
Low-middle 37 (50.68%) 40 (53.33%)
Poorly 10 (13.70%) 12 (16.00%)
Complications, n(%)
All Complication 53 (70.67%) 30 (40.00%) <0.001
Clavien-Dindo <0.001
No 20 (27.40%) 45 (60.00%)
Grade I-II 41 (56.16%) 22 (29.33%)
Grade III-V 12 (16.44%) 8 (10.67%)
Serious complications 12 (16.44%) 8 (10.67%) 0.432
Infectious Complication 21 (28.77%) 10 (13.33%) 0.035
Data are presented as n(%) or median (IQR). IQR, Interquartile range.
He et al. 10.3389/fsurg.2022.1091534
Frontiers in Surgery 07 frontiersin.org
indicator for various cancer patients (21–24). Some previous
studies have found that a low PNI is positively correlated with
poor prognosis in patients with biliary tract cancer, and the
use of this parameter has shown the potential to improve
prediction and identify high-risk patients more accurately and
precisely (25). A meta-analysis found that PNI was an
independent prognostic factor for overall survival and
postoperative complications in cancer patients. In addition,
subgroups with cutoff values less than 45 reported higher OR
values, which indirectly confirmed that a low PNI was indeed
associated with poor prognosis in some cancers (26).
However, this value has not been well-studied in patients
with dCCA. We compared the correlation and discrimination
between the INIs and complications and selected the PNI to
distinguish between malnourished and well-nourished
patients. Malnutrition is associated with an increased risk of
perioperative morbidity and mortality (27) and is a risk factor
for postoperative complications (28). A valid and objective
preoperative nutritional risk assessment is of great benefitto
both clinicians and patients, helping clinical doctors establish
the corresponding diagnosis and treatment strategy to achieve
improved clinical outcomes. The mechanism by which PNI
affects complications after radical pancreaticoduodenectomy
for dCCA is not very clear and may be related to albumin
and lymphocyte levels, thus reflecting the nutritional and
immune status of patients.
TABLE 4 Logistic regression analysis to identify risk factors for overall complications.
Characteristics Univariate analysis Multivariate analysis
OR 95%CI PValue OR 95%CI PValue
Age (years) 1.02 0.99–1.06 0.239
Sex 0.66 0.34–1.28 0.219
Prognostic nutritional index 0.85 0.79–0.93 <0.001 0.87 0.80–0.94 0.001
Leukocyte (10
9
/L) 1.28 1.09–1.49 0.002 1.19 1.01–1.44 0.052
Erythrocyte (10
9
/L) 0.70 0.39–1.27 0.245
Hemoglobin (g/L) 1.00 0.98–1.01 0.658
Aspartate transaminase (U/L) 1.00 0.99–1.01 0.992
Alanine transaminase (U/L) 1.00 0.99–1.01 0.389
Total bilirubin (µmol/L) 1.00 1–1.01 0.001 1.01 1.00–1.01 0.017
Total protein (g/L) 0.99 0.94–1.03 0.553
Cholesterol (mmol/L) 1.00 0.88–1.14 0.993
Uric acid (µmol/L) 1.00 1–1.01 0.540
Glucose (mm/L) 1.01 0.88–1.15 0.896
Triglycerides (mm/L) 1.51 1.04–2.19 0.028 1.24 0.79–1.94 0.349
Prothrombin time (S) 1.34 1.06–1.70 0.014 1.10 0.85–1.50 0.478
International normalized ratio 1.59 0.49–5.19 0.441
Carbohydrate antigen 199 (U/ml) 1.00 0.99–1.01 0.792
Carcinoembryonic antigen (ng/MI) 0.98 0.94–1.03 0.521
FIGURE 4
Prediction model for overall complication using CART analysis.
He et al. 10.3389/fsurg.2022.1091534
Frontiers in Surgery 08 frontiersin.org
The TBIL level can be used to assess the degree of jaundice
and hepatobiliary damage. Hyperbilirubinemia is considered to
be a potential high-risk factor associated with postoperative
mortality in hilar cholangiocarcinoma (29). A recent
retrospective, single-center study once again confirmed that
preoperative bilirubin concentrations were an important risk
factor for postoperative severe complications and mortality in
perihilar cholangiocarcinoma, with optimal cutoff values of
219.23 µmol/L and 548.08 µmol/L, respectively (30). A
multicenter European study also showed that a high
preoperative TBIL (≥265.2 µmol/L) was significantly
associated with increased complications after the major
resection of perihilar cholangiocarcinoma (31). Jaundice is
associated with immune dysfunction, increased bacterial
translocation, and deterioration of nutritional status and liver
function (32).
Normally, bilirubin levels need to be less than twice the
upper limit of normal before surgery, and preoperative biliary
drainage is recommended for patients requiring
decompression (TBIL > 250 µmol/L); otherwise, preoperative
biliary drainage should be avoided (33). We do not have level
1 evidence for individuals with high serum bilirubin levels.
Based on this study, preoperative jaundice reduction is
recommended for patients with a TBIL > 194.1 µmol/L to
improve performance status and survival, and preoperative
biliary drainage is recommended for patients with high
bilirubin (TBIL > 250 µmol/L) who require decompression.
Preoperative nutritional treatment is increasingly
recognized as an important part of surgical care. Nearly 50%
of hospitalized patients are malnourished or at risk of
malnutrition, and hospitalized and surgical patients who are
malnourished have significantly worse clinical outcomes (34).
Screening for malnutrition before major surgery is essential
because it can identify patients at risk of malnutrition who
may benefit from preoperative nutritional intervention (35),
which can improve patient outcomes (36). This study shows
that preoperative PNI and TBIL levels were independent risk
factors for short-term postoperative complications of dCCA.
Based on the CART analysis, a high complication rate
(90.24%) was found in patients with PNI ≤46.9 and TBIL >
194.1 µmol/L, which was useful for us to distinguish patients
at potentially high risk preoperatively. Therefore, preoperative
evaluations of patient nutritional status and TBIL level to help
guide clinicians in formulating corresponding diagnoses and
treatment strategies are expected to reduce the incidence of
postoperative complications, improve quality of life, and
improve the long-term prognosis of patients with dCCA.
This study has some limitations. Since the study was a
retrospective analysis of data from a single center, these
results may be affected by selection bias, and no further
follow-up visit, more prospective, large-sample, multicenter
studies are needed for further validation.
Conclusions
Nutritional assessments are necessary before radical
pancreaticoduodenectomy for dCCA. Preoperative PNI and
TBIL levels can be used as predictive markers for the risk of
postoperative complications. Effective perioperative intervention
for patients with low PNI and high TBIL levels can further
improve the surgical outcomes of patients with dCCA.
FIGURE 5
ROC curve of the decision tree.
TABLE 5 Relationship between risk group established by the
prediction model and complications.
Complications High-risk
group
(n= 41)
Middle-
risk group
(n= 32)
Low-risk
group
(n= 75)
P
value
All complications <0.001
Yes (n= 83) 37 16 30
No (n= 65) 4 16 45
CD Grades I-II <0.001
Yes (n= 63) 28 13 22
No (n= 85) 13 19 53
CD Grades III-V
Yes (n= 20) 9 3 8 0.175
No (n= 128) 32 29 67
Data are presented as n(%). CD, Clavien–Dindo classification.
He et al. 10.3389/fsurg.2022.1091534
Frontiers in Surgery 09 frontiersin.org
Data availability statement
The raw data supporting the conclusions of this article will
be made available by the authors, without undue reservation.
Ethics statement
Written informed consent was obtained from the individual
(s) for the publication of any potentially identifiable images or
data included in this article.
Author contributions
YLH and HRL make the same contribution to this work.
Correspondence Auther: PY and WBM make the same
contribution to this work. BB, YZ, YYL, PY, and WBM:
constructing an idea, formulating research objectives, and
research implications, developing a proposal, and critically
revising the content of the manuscript. YLH, HRL, YYM, and
JLL: researching literature, collecting data, collating data, and
analyzing data. JDZ, YXR, CLD, BB, YZ, YYL, PY, and WBM:
environmental support, providing relevant research tools and
software, taking responsible for pathology, revising the
content of the manuscript before submission. YLH, HRL, PY,
and WBM: Taking responsibility for the raw data, taking
responsibility for the logical interpretation, and presentation
of the results, and deciding to submit the manuscript for
publication. All authors contributed to the article and
approved the submitted version.
Funding
This work was supported by the National Natural Science
Foundation of China (Grant No. 82060551); The Foundation
of The First Hospital of Lanzhou University (Grant No.
ldyyyn2019-20); Natural Science Foundation of Gansu
Province (Grant No. 20JR10RA676); Science and Technology
Planning Project of Chengguan District in Lanzhou (Grant
No. 2020JSCX0043) and Natural Science Foundation of Gansu
Province (Grant No. 20JR10FA703).
Conflict of interest
The authors declare that the research was conducted in the
absence of any commercial or financial relationships that could
be construed as a potential conflict of interest.
Publisher’s note
All claims expressed in this article are solely those of the
authors and do not necessarily represent those of their
affiliated organizations, or those of the publisher, the editors
and the reviewers. Any product that may be evaluated in this
article, or claim that may be made by its manufacturer, is not
guaranteed or endorsed by the publisher.
Supplementary material
The Supplementary Material for this article can be found
online at: https://www.frontiersin.org/articles/10.3389/fsurg.
2022.1091534/full#supplementary-material
References
1. Brindley PJ, Bachini M, Ilyas SI, Khan SA, Loukas A, Sirica AE, et al.
Cholangiocarcinoma. Nat Rev Dis Primers. (2021) 7:65. doi: 10.1038/s41572-
021-00300-2
2. Rizvi S, Khan SA, Hallemeier CL, Kelley RK, Gores GJ. Cholangiocarcinoma -
evolving concepts and therapeutic strategies. Nat Rev Clin Oncol. (2018)
15:95–111. doi: 10.1038/nrclinonc.2017.157
3. Bahra M. Operative therapie distaler cholangiokarzinome [surgical treatment
of distal cholangiocarcinoma]. Chirurg. (2021) 92:788–95. doi: 10.1007/s00104-
021-01453-2
4. Lee RM, Maithel SK. Approaches and outcomes to distal cholangiocarcinoma.
Surg Oncol Clin N Am. (2019) 28:631–43. doi: 10.1016/j.soc.2019.06.014
5. Chen Q, Zheng Y, Chen J, Zhou J, Zhao J, Bi X, et al. Differences of
intraoperative outcomes and postoperative complications between intrahepatic
cholangiocarcinoma and colorectal liver metastasis in different surgical
methods. Transl Cancer Res. (2021) 10:4020–32. doi: 10.21037/tcr-21-553
6. Miyata T, Yamashita YI, Yamao T, Umezaki N, Tsukamoto M, Kitano Y, et al.
Prognostic impacts of postoperative complications in patients with intrahepatic
cholangiocarcinoma after curative operations. Int J Clin Oncol. (2017)
22:526–32. doi: 10.1007/s10147-017-1099-9
7. Virizuela JA, Camblor-Álvarez M, Luengo-Pérez LM, Grande E, Álvarez-
Hernández J, Sendrós-Madroño MJ, et al. Nutritional support and parenteral
nutrition in cancer patients: an expert consensus report. Clin Transl Oncol.
(2018) 20:619–29. doi: 10.1007/s12094-017-1757-4
8. European Association for the Study of the Liver. Electronic address:
easloffice@easloffice.eu; European Association for the Study of the Liver. EASL
Clinical practice guidelines on nutrition in chronic liver disease. J Hepatol.
(2019) 70:172–93. doi: 10.1016/j.jhep.2018.06.024
9. Chen L, Bai P, Kong X, Huang S, Wang Z, Wang X, Fang Y, et al. Prognostic
nutritional Index (PNI) in patients with breast cancer treated with neoadjuvant
chemotherapy as a useful prognostic indicator. Front Cell Dev Biol. (2021)
9:656741. doi: 10.3389/fcell.2021.656741
10. Dindo D, Demartines N, Clavien PA. Classification of surgical complications:
a new proposal with evaluation in a cohort of 6336 patients and results of a survey.
Ann Surg. (2004) 240:205–13. doi: 10.1097/01.sla.0000133083.54934.ae
11. Li Y, Chen L, Xing C, Ding C, Zhang H, Wang S, et al. Changes in
Serum lactate level predict postoperative intra-abdominal infection after pancreatic
resection. World J Surg. (2021) 45:1877–86. doi: 10.1007/s00268-021-05987-8
12. Wente MN, Bassi C, Dervenis C, Fingerhut A, Gouma DJ, Izbicki JR, et al.
Delayed gastric emptying (DGE) after pancreatic surgery: a suggested definition by
He et al. 10.3389/fsurg.2022.1091534
Frontiers in Surgery 10 frontiersin.org
the international study group of pancreatic surgery (ISGPS). Surgery. (2007)
142:761–8. doi: 10.1016/j.surg.2007.05.005
13. Hosaka H, Takeuchi M, Imoto T, Yagishita H, Yu A, Maeda Y, et al.
Machine learning-based model for predicting postoperative complications
among patients with colonic perforation: a retrospective study. J Anus Rectum
Colon. (2021) 5:274–80. doi: 10.23922/jarc.2021-010
14. Zhou W, Qian L, Rong Y, Zhou Q, Shan J, Li P, et al. Prognostic factors and
patterns of recurrence after curative resection for patients with distal
cholangiocarcinoma. Radiother Oncol. (2020) 147:111–7. doi: 10.1016/j.radonc.
2020.03.017
15. Chua TC, Mittal A, Arena J, Sheen A, Gill AJ, Samra JS. Resection margin
influences survival after pancreatoduodenectomy for distal cholangiocarcinoma.
Am J Surg. (2017) 213:1072–6. doi: 10.1016/j.amjsurg.2016.09.049
16. Liu H, Cen X, Suo T, Cai X, Yuan X, Shen S, et al. Trends and hospital
variations in surgical outcomes for cholangiocarcinoma in New York state.
World J Surg. (2017) 41:525–37. doi: 10.1007/s00268-016-3733-5
17. Onodera T, Goseki N, Kosaki G. Prognostic nutritional index in
gastrointestinal surgery of malnourished cancer patients. Nihon Geka Gakkai
Zasshi. (1984) 85(9):1001–5. PMID: 6438478
18. Li D, Yuan X, Liu J, Li C, Li W. Prognostic value of prognostic nutritional
index in lung cancer: a meta-analysis. J Thorac Dis. (2018) 10:5298–307. doi: 10.
21037/jtd.2018.08.51
19. Zhang F, Liu Z, Liang J, Liu S, Wu K, Zhang F, et al. Association between
preoperative serum albumin and prognosis in patients with adrenocortical
carcinoma after primary resection: a retrospective study. BMC Cancer. (2021)
21:961. doi: 10.1186/s12885-021-08689-5
20. Feng F, Zheng G, Wang Q, Liu S, Liu Z, Xu G, et al. Low lymphocyte count
and high monocyte count predicts poor prognosis of gastric cancer. BMC
Gastroenterol. (2018) 18:148. doi: 10.1186/s12876-018-0877-9
21. Watanabe H, Yamada T, Komori K, Hara K, Kano K, Takahashi K, et al.
Effect of prognostic nutrition index in gastric or gastro-oesophageal junction
cancer patients undergoing nivolumab monotherapy. In Vivo. (2021) 35:563–9.
doi: 10.21873/invivo.12292
22. Kanda M. Preoperative predictors of postoperative complications after gastric
cancer resection. Surg Today. (2020) 50:3–11. doi: 10.1007/s00595-019-01877-8
23. Jiang AM, Zhao R, Liu N, Ma YY, Ren MD, Tian T, et al. The prognostic
value of pretreatment prognostic nutritional index in patients with small cell
lung cancer and it’sinfluencing factors: a meta-analysis of observational studies.
J Thorac Dis. (2020) 12:5718–28. doi: 10.21037/jtd-20-1739
24. Wang X, Wang Y. The prognostic nutritional index is prognostic factor of
gynecological cancer: a systematic review and meta-analysis. Int J Surg. (2019)
67:79–86. doi: 10.1016/j.ijsu.2019.05.018
25. Wang J, Bo X, Li M, Nan L, Wang C, Gao Z, et al. Prediction efficacy for
clinical outcome of prognostic nutritional index in patients with resectable
biliary tract cancer depends on sex and obstructive jaundice status. Ann Surg
Oncol. (2021) 28:430–8. doi: 10.1245/s10434-020-08728-8
26. Sun K, Chen S, Xu J, Li G, He Y. The prognostic significance of the
prognostic nutritional index in cancer: a systematic review and meta-analysis.
J Cancer Res Clin Oncol. (2014) 140:1537–49. doi: 10.1007/s00432-014-1714-3
27. Matsuda T, Umeda Y, Matsuda T, Endo Y, Sato D, Kojima T, et al.
Preoperative prognostic nutritional index predicts postoperative infectious
complications and oncological outcomes after hepatectomy in intrahepatic
cholangiocarcinoma. BMC Cancer. (2021) 21:708. doi: 10.1186/s12885-021-
08424-0
28. Benoist S, Brouquet A. Nutritional assessment and screening for
malnutrition. J Visc Surg. (2015) 152:S3–7. doi: 10.1016/S1878-7886(15)30003-5
29. Neuhaus H. Preoperative biliary drainage in hilar cholangiocarcinoma:
when and how? Endosc Int Open. (2020) 08:E211–3. doi: 10.1055/a-0990-9912
30. Wronka KM, Grat M, Stypulkowski J, Bik E, Patkowski W, Krawczyk M,
et al. Relevance of preoperative hyperbilirubinemia in patients undergoing
hepatobiliary resection for hilar cholangiocarcinoma. J Clin Med. (2019) 8:458.
doi: 10.3390/jcm8040458
31. Farges O, Regimbeau JM, Fuks D, Le Treut YP, Cherqui D, Bachellier P,
et al. Multicentre European study of preoperative biliary drainage for hilar
cholangiocarcinoma. Br J Surg. (2013) 100:274–83. doi: 10.1002/bjs.8950
32. Lleo A, Colapietro F, Maisonneuve P, Aloise M, Craviotto V, Ceriani R, et al.
Risk stratification of cholangiocarcinoma patients presenting with jaundice: a
retrospective analysis from a tertiary referral center. Cancers (Basel). (2021)
13:2070. doi: 10.3390/cancers13092070
33. Melloul E, Lassen K, Roulin D, Grass F, Perinel J, Adham M, et al.
Guidelines for perioperative care for pancreatoduodenectomy: Enhanced
Recovery After Surgery (ERAS) recommendations 2019. World J Surg. (2020)
44:2056–84. doi: 10.1007/s00268-020-05462-w
34. Gillis C, Wischmeyer PE. Pre-operative nutrition and the elective surgical
patient: why, how and what? Anaesthesia. (2019) 74:27–35. doi: 10.1111/anae.
14506
35. Wischmeyer PE, Carli F, Evans DC, Guilbert S, Kozar R, Pryor A, et al.
American Society for enhanced recovery and perioperative quality initiative
joint consensus statement on nutrition screening and therapy within a surgical
enhanced recovery pathway. Anesth Analg. (2018) 126:1883–95. doi: 10.1213/
ANE.0000000000002743
36. Wobith M, Weimann A. Oral nutritional supplements and enteral nutrition
in patients with gastrointestinal surgery. Nutrients. (2021) 13:2655. doi: 10.3390/
nu13082655
He et al. 10.3389/fsurg.2022.1091534
Frontiers in Surgery 11 frontiersin.org
Available via license: CC BY
Content may be subject to copyright.
