Abstract
Backgrounds/Aims
The role of preoperative biliary drainage (PBD) in improving perioperative outcomes for patients with obstructive jaundice remains controversial. This retrospective study aims to evaluate the impact of PBD in patients with varying severities of jaundice and to identify optimal strategies to minimize complications and mortality following pancreaticoduodenectomy (PD).
Methods
This retrospective clinical study included 240 patients with obstructive jaundice who underwent PD. Patients were categorized into severe Jaundice group (SJ group) and moderate Jaundice group (MJ group). Preoperative, intraoperative, and postoperative clinical data were compared based on whether PBD was performed. Additionally, the association between PBD duration and postoperative complications and mortality after PD was assessed.
Results
Among 115 patients in the SJ group, 94 received PBD, whereas 46 of the 125 patients in the MJ group received PBD. In the SJ-PBD group, the rate of postoperative bile leakage was significantly lower compared with the direct surgery group; however, overall complication rates did not differ. In the MJ-PBD group, the incidence of incision-related complications increased significantly, with no notable reduction in overall postoperative complications. Patients who underwent PBD for at least 10 days experienced lower rates of severe complications (Clavien-Dindo grade ≥ 3) than patients with PBD duration less than 10 days (14.3% vs. 25.5%, p = 0.012).
Conclusions
PBD should be considered in obstructive jaundiced patients with total bilirubin > 256 µmol/L, particularly those presenting with systemic functional impairment. A minimum PBD duration of 10 days appears beneficial for reducing the risk of serious postoperative complications and mortality.
Keywords: Surgical procedure, Pancreaticoduodenectomy, Obstructive jaundice, Postoperative complications, Postoperative mortality
INTRODUCTION
Obstructive jaundice frequently complicates the management of patients scheduled for curative pancreaticoduodenectomy (PD) for resectable periampullary cancers, including pancreatic ductal adenocarcinoma, ampullary carcinoma, and biliary carcinoma. Approximately 50% to 80% of patients diagnosed with periampullary adenocarcinoma present with obstructive jaundice [1]. The necessity of preoperative biliary drainage (PBD) in these patients remains under debate. Despite numerous clinical reports, no universally accepted professional guidelines have been established to guide PBD decision-making in this context [2]. Current evidence suggests that routine PBD prior to PD does not reduce postoperative complications or enhance prognosis, but is instead associated with increased rates of infectious complications, surgical site infection, and delayed gastric emptying (DGE) [3].
Theoretically, patients with jaundice may present with abnormalities in biochemistry, immune function, and various organ systems due to cholestasis, impaired hepatic and renal function, altered bilirubin metabolism, deficient fat digestion and absorption, increased susceptibility to infection, coagulation disturbances, and systemic complications at the time they seek hospital care. Therefore, safely performing immediate radical PD during the perioperative period remains challenging. A meta-analysis published in 2018 [4] demonstrated that the consequent systemic inflammatory response syndrome can progress to multiple organ dysfunction syndrome by affecting proinflammatory mediators, coagulation profiles, hemodynamics, and organ function in cases of obstructive jaundice. Optimal perioperative management—including correction of water and electrolyte imbalances, administration of vitamin K, and prophylactic antibiotics—remains crucial for improving patient outcomes. However, PBD itself is associated with potential risks such as cholangitis, pancreatitis, and hemobilia, and may increase specific postoperative complications [5]. Several statistical analyses utilizing clinical data have indicated that PBD could elevate rates of complications like DGE and surgical site infection, without decreasing overall complication rates or mortality following PD [6]. As such, routine PBD is not recommended. Nonetheless, for patients with periampullary malignancy who require surgery to be postponed, PBD is considered a safe and effective preoperative bridging strategy [7]. Additionally, some experts recommend that patients with serum total bilirubin (TB) levels exceeding 250 μmol/L undergo routine PBD prior to pancreatectomy to minimize the risks of postoperative hemorrhage and pancreatic fistula [8]. Selection bias exists among surgical centers of various case volumes, leading to disparate conclusions, but ongoing research continues to refine the indications for PBD to guide surgical decision-making in these patients.
In this study, we retrospectively collected clinical data from all patients undergoing PD at our hospital between February 2015 and December 2023, aiming to assess the impact of direct surgery (DS) compared to PBD on postoperative complications, in-hospital mortality, and 90-day mortality in patients with varying levels of jaundice to further evaluate the appropriateness of PBD.
MATERIALS AND METHODS
We performed a retrospective cohort study at a single center, enrolling patients who underwent curative PD—including Whipple’s operation and pylorus-preserving pancreaticoduodenectomy (PPPD)—with or without PBD in the Department of Hepato-Pancreato-Biliary Surgery from February 2015 to December 2023. Ethical approval was obtained from the institutional Medical Ethics Review Committee, and the requirement for informed consent was waived due to the retrospective study design (P20240220-001).
Patients and grouping
A total of 532 patients with complete clinical and laboratory data who underwent PD at this institution were available and included in this retrospective study. Of these, 179 cases did not have preoperative obstructive jaundice, and 113 patients with preoperative TB elevation below 128 µmol/L underwent PD directly. The remaining 240 patients with obstructive jaundice included in this retrospective analysis were categorized into severe Jaundice group (SJ group) and moderate Jaundice group (MJ group) based on whether their maximum preoperative TB exceeded 256 µmol/L. Additionally, these patients were allocated to either the PBD group or DS group according to whether PBD was performed.
PBD and operative procedure
All patients undergoing PD, with or without PBD, were clinically diagnosed with benign or malignant tumors in areas such as the pancreatic head, ampulla, distal common bile duct, or duodenum. They all underwent comprehensive physical and laboratory evaluations, and those without obvious contraindications participated in a multidisciplinary preoperative discussion confirming the appropriateness of surgery. At our medical center, if there was a clear indication for PD and all preoperative preparations were complete, patients with a maximum TB less than 128 µmol/L were not considered for PBD and proceeded directly to surgery. For jaundiced patients with TB levels greater than 128 µmol/L, if coagulation tests, nutritional status, and overall physical condition were satisfactory, DS was often selected over PBD by the surgical team. The criteria for proceeding with PBD specified that patients should both be likely to benefit from PD and have a preoperative maximum TB greater than 128 µmol/L. This TB threshold was a necessary condition for PBD, and for patients meeting these criteria, the attending physician, considering individual circumstances, determined whether PBD was appropriate. Patients who had received PBD in other departments or hospitals, and subsequently underwent surgery in our department, were also included in the analysis if their clinical data were complete. Individuals who underwent T-tube drainage surgery for non-tumor-related reasons, or who received neoadjuvant chemotherapy after PBD, were excluded from the study. The main types of PBD performed included stent implantation, endoscopic nasobiliary drainage (ENBD), and percutaneous transhepatic cholangial drainage (PTCD). PTCD was most commonly utilized in our institution, and timing of surgery following PBD depended on the surgeon’s discretion, typically occurring more than 4 days later, by which time TB would drop below 128 µmol/L. Perioperative antibiotics were used as appropriate. The decision between PPPD and Whipple was based on the attending doctor’s clinical judgment and patient condition. Regardless of the pancreaticojejunostomy technique, a pancreatic duct stent was always inserted. All procedures were performed by highly experienced surgeons.
Data collection and definitions
Preoperative clinical data collected included age, sex, body mass index (BMI), hypertension, diabetes mellitus, maximum preoperative TB, last TB prior to surgery, hypoalbuminemia within 30 days before surgery (HALB30), and presence of anemia in the 30 days before surgery (AN30). Data also encompassed whether PBD was performed preoperatively and the type and duration of drainage. Intraoperative clinical data comprised surgical approach, duration of surgery, intraoperative blood loss, intraoperative transfusion, and performance of combined excision involving other organs or tissues. Postoperative clinical data included the occurrence of postoperative pancreatic fistula (POPF), post-pancreatectomy hemorrhage (PPH), biliary fistula, chylous fistula, anastomotic leakage, DGE, intra-abdominal infection, incision-related complication, pulmonary complications, postoperative blood transfusion, postoperative invasive interventions, re-intubation, reoperation, total length of hospital stay, postoperative duration of stay, and postoperative pathological outcomes. Overall postoperative complications were assessed using the Clavien-Dindo classification (CD) [9,10]. In-hospital mortality and deaths occurring within 90 days after surgery were also incorporated into postoperative clinical outcomes. The definitions of POPF, PPH, biliary fistula, and DGE were based on the criteria established by the International Study Group in Pancreatic Surgery (ISGPS) [11-14].
Statistical analysis
Skewed continuous variables are presented as median and interquartile range. Continuous variables exhibiting normal or near-normal distribution are expressed as mean ± standard deviation, while categorical variables are presented as frequency (%). Comparisons of continuous data between groups utilized t-tests, and categorical data were analyzed by either chi-square or Fisher’s exact tests as appropriate. All analyses were performed using IBM SPSS version 26 (IBM Corp.). Results with p-values <0.05 were considered statistically significant.
RESULTS
Of the 532 PDs conducted in our center between February 1, 2015 and December 31, 2023, 140 patients (94 in the SJ-PBD group and 46 in the MJ-PBD group) received PBD prior to surgery. Among these patients, the lowest recorded preoperative maximum TB was 131.7 µmol/L. As a result, we analyzed data from 100 patients (21 in the SJ-DS group and 79 in the MJ-DS group) who underwent DS with preoperative maximum TB greater than 128 µmol/L for comparison.
The characteristics of the patients are presented in Table 1. There were no significant differences in age, sex, BMI, comorbidities, preoperative TB, surgical method, duration of operation, blood loss volume, or intraoperative blood transfusion rate among the groups. However, preoperative hypoalbuminemia and anemia were more frequent in the PBD group, particularly in the MJ-PBD group (hypoalbumin, p = 0.006; anemia, p = 0.002). PTCD constituted the primary method of PBD (SJ-PBD, 92.6%; MJ-PBD, 95.7%). Pathological findings revealed that 118 (49.2%) patients were diagnosed with pancreatic adenocarcinoma, 52 (21.7%) with ampullary carcinoma, 47 (19.6%) with distal cholangiocarcinoma, and 17 (7.1%) with duodenal adenocarcinoma.
Table 1.
Basic demographic and clinical characteristics of the patients in the study
| Variable | SJ group | X2/t value | p-value | MJ group | X2/t value | p-value | ||
|---|---|---|---|---|---|---|---|---|
| PBD group (n = 94) |
DS group (n = 21) |
PBD group (n = 46) |
DS group (n = 79) |
|||||
| Age (yr)a) | 65.19 [8.954] | 63.24 [10.183] | 0.881 | 0.38 | 64.78 [8.947] | 65.39 [9.862] | –0.345 | 0.731 |
| Maleb) | 65 (69.1) | 12 (57.1) | 1.118 | 0.29 | 28 (60.9) | 49 (62.0) | 0.016 | 0.898 |
| BMI (kg/m2)a,c) | 22.5 (20.8, 24.2) | 24.9 (23.1, 26.0) | –3.338 | 0.001 | 23.27 (3.627) | 23.43 (2.874) | –0.27 | 0.788 |
| Hypertensionb) | 39 (41.5) | 10 (47.6) | 0.264 | 0.608 | 15 (32.6) | 31 (39.2) | 0.55 | 0.458 |
| DMb,d) | 26 (27.7) | 4 (19.0) | 0.289 | 0.591 | 9 (19.6) | 10 (12.7) | 1.076 | 0.300 |
| Max-TB (µmol/L)a,c) | 344 (303.6, 392.63) | 283.9 (263.55, 332) | 3.612 | < 0.001 | 193.81 (37.289) | 175.08 (29.657) | 2.913 | 0.005 |
| Last-TB (µmol/L)a,c) | 107.3 (72.375, 170.2) | 283.9 (263.55, 332) | –6.790 | < 0.001 | 92.19 (44.4) | 164.42 (40.204) | –9.077 | < 0.001 |
| HALB30b) | 59 (62.8) | 9 (42.9) | 2.815 | 0.093 | 28 (60.9) | 28 (35.4) | 7.6 | 0.006 |
| AN30b,d) | 34 (36.2) | 4 (19.0) | 1.567 | 0.211 | 20 (43.5) | 14 (17.7) | 9.74 | 0.002 |
| PBD typee) | - | < 0.001 | - | < 0.001 | ||||
| PTCD | 87 (92.6) | 0 (0) | 44 (95.7) | 0 (0) | ||||
| ENBD | 3 (3.2) | 0 (0) | 0 (0) | 0 (0) | ||||
| Stents | 4 (4.3) | 0 (0) | 2 (4.3) | 0 (0) | ||||
| Surgical methodb) | 0.437 | 0.509 | 0.046 | 0.831 | ||||
| Whipple | 60 (63.8) | 15 (71.4) | 30 (65.2) | 53 (67.1) | ||||
| PPPD | 34 (36.2) | 6 (28.6) | 16 (34.8) | 26 (32.9) | ||||
| Surgical methode) | - | 0.583 | - | - | ||||
| Open | 89 (94.7) | 21 (100) | 46 (100) | 79 (100) | ||||
| Operating time (min)a) | 279.8 [73.449] | 295.52 [56.771] | –0.92 | 0.359 | 269.67 [62.538] | 281.44 [57.042] | –1.074 | 0.285 |
| Blood loss volume (mL)c) | 400 (300, 500) | 400 (200, 550) | –0.509 | 0.61 | 400 (187.5, 525) | 400 (200, 600) | –0.093 | 0.926 |
| Transfusionb) | 57 (60.6) | 13 (61.9) | 0.012 | 0.914 | 24 (52.2) | 44 (55.7) | 0.145 | 0.703 |
| Excision of other organd) | 2 (2.1) | 1 (4.8) | 0.401 | 0.527 | 3 (6.5) | 6 (7.6) | 0 | > 0.999 |
| Pathological results | - | 0.156 | - | 0.226 | ||||
| PDAC | 42 (44.7) | 9 (42.9) | 20 (43.5) | 47 (59.5) | ||||
| Ampullary carcinoma | 27 (28.7) | 5 (23.8) | 11 (23.9) | 9 (11.4) | ||||
| Biliary carcinoma | 20 (21.3) | 3 (14.3) | 11 (23.9) | 13 (16.5) | ||||
| Duodenal adenocarcinoma | 2 (2.1) | 4 (19.0) | 4 (8.7) | 7 (8.9) | ||||
| pNET | 1 (1.1) | 0 (0) | 0 (0) | 0 (0) | ||||
| IPMN | 0 (0) | 0 (0) | 0 (0) | 0 (0) | ||||
| Other malignant | 1 (1.1) | 0 (0) | 0 (0) | 1 (1.3) | ||||
| Other benign | 1 (1.1) | 0 (0) | 0 (0) | 2 (2.5) | ||||
Values are presented as mean [standard deviation], number (%), or median (interquartile range).
SJ group, severe Jaundice group; MJ group, moderate Jaundice group; BMI, body mass index; DM, diabetes mellitus; TB, total bilirubin; PBD, preoperative biliary drainage; PTCD, percutaneous transhepatic cholangial drainage; ENBD, endoscopic nasobiliary drainage; PPPD, pylorus-preserving pancreaticoduodenectomy; PDAC, pancreatic adenocarcinoma; pNET, pancreatic neuroectodermal tumor; IPMN, intraductal papillary mucinous neoplasm.
a)Student’s t-test was used to compare differences in continuous variables.
b)Chi-square test was used to compare differences in categorical variables.
c)Wilcoxon rank-sum test was used to compare differences in continuous variables.
d)Continuity-corrected chi-square test was used to compare differences in categorical variables.
e)Fisher’s exact test was used to compare differences in categorical variables. Note: Fisher’s exact test does not yield a test statistic.
Postoperative outcomes are displayed in Table 2. Based on varying levels of highest preoperative TB, major postoperative complications were independently assessed between the SJ-PBD group versus the SJ-DS group and the MJ-PBD group versus the MJ-DS group. The occurrence of POPF (Grade B and Grade C) in SJ-PBD and MJ-PBD was 22 (23.4%) and 5 (10.9%), respectively; however, there was no statistical significance in the incidence of pancreatic fistula between the respective DS groups (p = 0.263, p = 0.767). Although a higher proportion of PPH was observed in the SJ group compared to the MJ group, statistical comparison between the PBD and DS groups did not reveal significant differences (p = 0.290, p = 0.238). Notably, in the SJ subgroup, the incidence of bile leakage was significantly greater in the DS group versus the PBD group (38.1% vs. 13.8%, p = 0.022), whereas this difference was not significant in the MJ subgroup (p = 0.089). Furthermore, Grade A DGE was less common in the SJ-PBD and MJ-PBD groups than in the corresponding SJ-DS and MJ-DS groups, whereas Grades B and C DGE were more frequent in the PBD groups (p = 0.067, p = 0.182), but overall, there were no statistically significant differences in DGE incidence between the groups (p = 0.292, p = 0.545).
Table 2.
Postoperative outcomes
| Variable | SJ group | X2/t value | p-value | MJ group | X2/t value | p-value | ||
|---|---|---|---|---|---|---|---|---|
| PBD group (n = 94) |
DS group (n = 21) |
PBD group (n = 46) |
DS group (n = 79) |
|||||
| POPFa,b,c) | ||||||||
| B | 19 (20.2) | 2 (9.5) | - | 0.473 | 3 (6.5) | 9 (11.4) | 1.869 | 0.393 |
| C | 3 (2.1) | 0 (0) | 2 (4.3) | 1 (1.3) | ||||
| B & C | 22 (23.4) | 2 (9.5) | 1.25 | 0.263 | 5 (10.9) | 10 (12.7) | 0.088 | 0.767 |
| PPHb,c) | ||||||||
| A | 18 (19.1) | 8 (38.1) | 3.706 | 0.157 | 6 (13.0) | 13 (16.5) | 1.71 | 0.425 |
| B & C | 11 (11.7) | 1 (4.8) | 3 (6.5) | 10 (12.7) | ||||
| A & B & C | 29 (30.9) | 9 (42.9) | 1.118 | 0.290 | 9 (19.6) | 23 (29.1) | 1.392 | 0.238 |
| Bile leakageb,c) | ||||||||
| A | 7 (7.4) | 3 (14.3) | 6.4 | 0.041 | 2 (4.3) | 6 (7.6) | 3.327 | 0.189 |
| B & C | 6 (6.4) | 5 (23.8) | 2 (4.3) | 10 (12.7) | ||||
| A & B & C | 13 (13.8) | 8 (38.1) | 5.243 | 0.022 | 4 (8.7) | 16 (20.3) | 2.889 | 0.089 |
| Chylous fistulab) | 3 (3.2) | 2 (9.5) | 0.483 | 0.487 | 3 (6.5) | 5 (6.3) | 0 | > 0.999 |
| Anastomosis leakagea) | 1 (1.1) | 0 (0) | - | 0.524 | 1 (2.2) | 0 (0) | - | 0.156 |
| DGEb, c) | ||||||||
| A | 12 (12.8) | 7 (33.3) | 5.405 | 0.067 | 5 (10.9) | 17 (21.5) | 3.408 | 0.182 |
| B & C | 13 (13.8) | 1 (4.8) | 5 (10.9) | 4 (5.1) | ||||
| A & B & C | 25 (26.6) | 8 (38.1) | 1.109 | 0.292 | 10 (21.7) | 21 (26.6) | 0.366 | 0.545 |
| Intra-abdominal infectionc) | 21 (22.3) | 6 (28.6) | 0.371 | 0.542 | 9 (19.6) | 8 (10.1) | 2.204 | 0.138 |
| Incision complicationb) | 12 (12.8) | 1 (4.8) | 0.444 | 0.505 | 10 (21.7) | 4 (5.1) | 6.538 | 0.011 |
| Pulmonary complicationsb,c) | 14 (14.9) | 3 (14.3) | 0 | > 0.999 | 7 (15.2) | 9 (11.4) | 0.381 | 0.537 |
| Postoperative blood transfusionc) | 29 (30.9) | 9 (42.9) | 1.118 | 0.29 | 9 (19.6) | 23 (29.1) | 1.392 | 0.238 |
| Postoperative invasive procedureb,c) | 19 (20.2) | 2 (9.5) | 0.695 | 0.404 | 7 (15.2) | 8 (10.1) | 0.713 | 0.398 |
| Re-intubationb) | 7 (7.4) | 1 (4.8) | 0 | > 0.999 | 2 (4.3) | 5 (6.3) | 0.004 | 0.951 |
| Re-surgerya,b) | 5 (5.3) | 0 (0) | - | 0.151 | 1 (2.2) | 3 (3.8) | 0 | > 0.999 |
| Postoperative hospital stays (day)d) | 16 (14, 20) | 17 (13, 20.5) | 0.160 | 0.873 | 16.5 (13, 21.25) | 16 (14, 19) | –1.14 | 0.254 |
| Length of hospital stay (day)e) | 30.394 [9.546] | 23.905 [7.422] | 2.92 | 0.004 | 29.57 [7.25] | 24.63 [7.03] | 3.739 | < 0.001 |
| Death in hospitala,b) | 4 (4.3) | 0 (0) | - | 0.2 | 2 (4.3) | 4 (5.1) | 0 | > 0.999 |
| Death in 90 daysa,b) | 4 (4.3) | 0 (0) | - | 0.2 | 3 (6.5) | 4 (5.1) | 0 | > 0.999 |
Values are presented as number (%), median (interquartile range), or mean [standard deviation].
SJ group, severe Jaundice group; MJ group, moderate Jaundice group; PBD, preoperative biliary drainage; DS, direct surgery group; POPF, postoperative pancreatic fistula; PPH, post-pancreatectomy hemorrhage; DGE, delayed gastric emptying.
a)Fisher’s exact test was used to compare differences in categorical variables. Note: Fisher’s exact test does not yield a test statistic.
b)Continuity-corrected chi-square test was used to compare differences in categorical variables.
c)Chi-square test was used to compare differences in categorical variables.
d)Wilcoxon rank-sum test was used to compare differences in continuous variables.
e)Student’s t-test was used to compare differences in continuous variables.
There was no significant difference between the groups, regardless of the severity of jaundice, in terms of other postoperative complications such as chylous fistula, anastomotic leak, intra-abdominal infection, pulmonary complications, or postoperative metrics including blood transfusion, invasive interventions, re-intubation, reoperation, and length of hospital stay (p > 0.05 for all). Additionally, as a result of prolonged drainage, the total hospital stay in the PBD group was significantly longer than that of the DS group, as anticipated.
Among all patients, 97 (69.3%) in the PBD group and 79 (79.0%) in the DS group experienced at least one mild postoperative complication and were classified as CD grade 1 or above, respectively. Although the rates of CD grades 1–2 in the PBD group were lower, as indicated in Table 3, 4, the proportion of severe complications (CD grades ≥ 3) in the PBD group was higher compared to the DS group (19.3% vs. 10.0%).
Table 3.
Overall postoperative complications and mortality
| PBD group (n = 140) |
DS group (n = 100) |
X2-value | p-value | |
|---|---|---|---|---|
| CDa) | 3.857 | 0.05 | ||
| < 3 | 113 (80.7) | 90 (90.0) | ||
| ≥ 3 | 27 (19.3) | 10 (10.0) | ||
| Death in hospitalb) | 6 (4.3) | 4 (4.0) | 0 | > 0.999 |
| Death in 90 daysb) | 7 (5.0) | 4 (4.0) | 0.003 | 0.958 |
Values are presented as number (%).
PBD, preoperative biliary drainage; DS, direct surgery group; CD, Clavien-Dindo classification.
a)Chi-square test was used to compare differences in categorical variables.
b)Continuity-corrected chi-square test was used to compare differences in categorical variables.
Table 4.
Overall postoperative complications and mortality
| SJ group | X2/t value | p-value | MJ group | X2/t value | p-value | |||
|---|---|---|---|---|---|---|---|---|
| PBD group (n = 94) |
DS group (n = 21) |
PBD group (n = 46) |
DS group (n = 79) |
|||||
| CDa,b) | 0.867 | 0.352 | 0.713 | 0.398 | ||||
| < 3 | 74 (78.7) | 19 (90.5) | 39 (84.8) | 71 (89.9) | ||||
| ≥ 3 | 20 (21.3) | 2 (9.5) | 7 (15.2) | 8 (10.1) | ||||
| Death in hospitalb,c) | 4 (4.3) | 0 (0) | - | 0.441 | 2 (4.3) | 4 (5.1) | 0 | > 0.999 |
| Death in 90 daysb,c) | 4 (4.3) | 0 (0) | - | 0.441 | 3 (6.5) | 4 (5.1) | 0 | > 0.999 |
Values are presented as number (%).
SJ group, severe Jaundice group; MJ group, moderate Jaundice group; PBD, preoperative biliary drainage; DS, direct surgery group; CD, Clavien-Dindo classification.
a)Chi-square test was used to compare differences in categorical variables.
b)Continuity-corrected chi-square test was used to compare differences in categorical variables.
c)Fisher’s exact test was used to compare differences in categorical variables. Note: Fisher’s exact test does not yield a test statistic.
The majority of patients in the PBD group received PTCD, with drainage durations before PD ranging from 4 days to 88 days. To assess the impact of PBD duration on postoperative complications and to determine the optimal timing for PBD, boxplots were employed to investigate the relationship between PBD duration and CD grade, as depicted in Fig. 1. The boxplot demonstrated that severe complications were more likely when the PTCD duration was less than 10 days. Group classification and statistical analysis are detailed in Table 5. Patients who underwent surgery following continuous drainage for more than 4 days but less than 10 days after achieving jaundice reduction with PTCD exhibited a higher incidence of severe postoperative complications (CD grade ≥ 3) and perioperative mortality than those who underwent PTCD for more than 10 days, and this difference was statistically significant (p = 0.012).
Fig. 1.
Relationship between preoperative biliary drainage duration and CD grade showed by boxplots. CD, Clavien-Dindo classification; PTCD, percutaneous transhepatic cholangial drainage.
Table 5.
Effect of duration of PTCD on severe complications and mortality
| PTCD | X2-value | p-value | ||
|---|---|---|---|---|
| < 10 days (n = 47) |
≥ 10 days (n = 84) |
|||
| CD ≥3a) | 12 (25.5) | 12 (14.3) | 10.976 | 0.012 |
| Death in hospitalb) | 3 (6.4) | 1 (1.2) | 1.271 | 0.260 |
| Death in 90 daysb) | 4 (8.5) | 1 (1.2) | 2.631 | 0.105 |
Values are presented as number (%).
PTCD, percutaneous transhepatic cholangial drainage; CD, Clavien-Dindo classification.
a)Chi-square test was used to compare differences in categorical variables.
b)Continuity-corrected chi-square test was used to compare differences in categorical variables.
Due to the significant findings presented in Table 5, major postoperative complications in groups with PBD duration shorter than 10 days and those with longer than 10 days were specifically summarized in Table 6 to explore potential explanations for the increased incidence of severe complications associated with PBD durations of less than 10 days prior to PD. The results showed that the rates of PPH (Grade B and C), postoperative blood transfusions, abdominal infection, and pulmonary infection were all higher in the group with PBD durations of up to 10 days prior to PD than in the longer duration group, although these differences did not reach statistical significance (p > 0.05 for all).
Table 6.
The detailed type of postoperative complication with having performed PBD within or more than 10 days of surgery
| PTCD | X2-value | p-value | ||
|---|---|---|---|---|
| PTCD < 10 days (n = 47) | PTCD ≥ 10 days (n = 84) | |||
| POPFa,b) | ||||
| B | 7 (14.9) | 14 (16.7) | 1.236 | 0.539 |
| C | 3 (6.4) | 1 (1.2) | ||
| B & C | 10 (21.3) | 15 (17.9) | 0.034 | 0.855 |
| PPHa) | 0.569 | 0.451 | ||
| B & C | 6 (12.8) | 6 (7.1) | ||
| Bile leakagea) | 0 | > 0.999 | ||
| B & C | 3 (6.4) | 5 (6.0) | ||
| DGEb) | 0.021 | 0.885 | ||
| B & C | 6 (12.8) | 10 (11.9) | ||
| Chylous fistulaa) | 3 (6.4) | 2 (2.4) | 0.451 | 0.502 |
| Anastomosis leakagec) | 1 (2.1) | 1 (1.2) | - | > 0.999 |
| Intra-abdominal infectionb) | 12 (25.5) | 15 (17.9) | 1.085 | 0.298 |
| Incision complicationb) | 6 (12.8) | 13 (15.5) | 0.354 | 0.552 |
| Pulmonary complicationsb) | 9 (19.1) | 8 (9.5) | 2.472 | 0.116 |
| Postoperative blood transfusionb) | 15 (31.9) | 20 (23.8) | 1.011 | 0.315 |
| Postoperative hospital stays (day)d) | 16 [10] | 15.50 [7] | –0.961 | 0.337 |
Values are presented as number (%) or median [interquartile range].
PTCD, percutaneous transhepatic cholangial drainage; POPF, postoperative pancreatic fistula; PPH, post-pancreatectomy hemorrhage; DGE, delayed gastric emptying.
a)Continuity-corrected chi-square test was used to compare differences in categorical variables.
b)Chi-square test was used to compare differences in categorical variables.
c)Fisher’s exact test was used to compare differences in categorical variables. Note: Fisher’s exact test does not yield a test statistic.
d)Wilcoxon rank-sum test was used to compare differences in continuous variables.
DISCUSSION
Preoperative obstructive jaundice can significantly influence the occurrence and progression of postoperative complications following PD. Consequently, there is ongoing debate regarding whether to perform PBD and under what specific conditions it should be implemented [15]. In reality, the impact of varying degrees of obstructive jaundice on perioperative general health and surgery-related complications differs considerably [16]. Multiple studies, including retrospective, prospective, and randomized controlled trials, have concluded that patients with obstructive jaundice undergoing PD do not derive benefit from routine PBD [17,18]. Our retrospective analysis produced results that contrast with these findings. In this study, we observed that when the TB levels of patients with preoperative obstructive jaundice were below 256 µmol/L, PBD was associated with an increased rate of incision infection and prolonged the overall hospital stay, without improving postoperative complications following PD. Conversely, when preoperative TB exceeded 256 µmol/L, although PBD played a partial role in reducing the incidence of postoperative bile leakage, it did not significantly improve rates of POPF (Grade B and C), PPH (Grade B and C), or severe complications (CD ≥ 3). However, the extended duration of disease and worsening patient status might have introduced selection bias into this retrospective analysis, influenced by clinical decision-making. Patients with severe hyperbilirubinemia experienced more prolonged obstructive jaundice, potentially worsening endotoxemia and nutritional status, and presenting greater challenges in tumor and regional lymph node removal. This suggests that patients in the PBD group may have had more complex or severe conditions, contributing to worse outcomes. As a result, surgeons might be more likely to delay surgery to optimize the patient’s status. Nonetheless, PBD appeared to offer certain benefits in reducing the rates of PPH and overall complications compared to the DS group. Pattarapuntakul et al. [15] reported that PBD reduces the risk of intraoperative bleeding, lowers postoperative complication rates, and may improve survival outcomes in patients with SJ. However, the incidence of severe complications (CD grade ≥ 3) and perioperative mortality was marginally higher in the PBD group than in the DS group, though this difference was not statistically significant. This may be attributed to poorer nutritional status, compromised liver function, anemia, and more advanced disease stages in the PBD group, rather than a direct increase in risk caused by PBD. Johansen et al. [19] have noted that PBD does not independently influence patient survival following PD, and even mildly elevated preoperative bilirubin levels may negatively affect long-term survival. A study including 33,516 patients from 39 retrospective cohort studies demonstrated similar findings, indicating that PBD should be considered for patients with severe infections or progressive jaundice, and that optimizing the duration and timing of PBD may reduce complications [20].
With respect to the appropriate choice of PBD type, many centers employ preoperative biliary stent implantation. It is important to highlight that numerous studies have shown a significant association between biliary stent placement and both overall and infectious complications [21]; furthermore, in certain situations, local edema may occur after stent insertion due to technical factors, creating additional challenges during surgery. In light of these findings and previous clinical experiences, most patients at our center underwent PTCD, which remains a feasible strategy for those preparing for PD [22]. These patients may experience a higher therapeutic success rate, and reduced risks of cholangitis and postoperative complications compared to those undergoing endoscopic biliary drainage, while also allowing for correction of hypoalbuminemia, anemia, and electrolyte imbalances during the preoperative drainage period [23,24]. Nevertheless, a multicenter retrospective analysis [25] suggested that PTCD may contribute to an increased risk of diffuse recurrence in biliary tract tumors, although, for pancreatic cancer, it is preferred due to fewer complications without altering recurrence rates. In situations where immediate surgery is not feasible, biliary stenting can be used to maintain nutritional status, which may influence survival outcomes [26].
Currently, there is no universally accepted consensus or definitive conclusion regarding the optimal duration of PBD in patients with obstructive jaundice. Twelve studies have compared short versus prolonged drainage periods following PBD, and results indicate no significant differences in mortality or major postoperative complications between the short and prolonged duration groups, suggesting that surgery can be performed at the earliest opportunity once optimal perioperative management is achieved [27]. Cao et al. [28] reported that a PBD duration of up to 3 weeks appears sufficient, while extending drainage beyond 6 weeks does not yield additional benefits. Nonetheless, other studies have found that the total incidence of postoperative complications correlates with the duration of PBD. Research by Zhu et al. [29] indicated that restricting drainage duration to within 2 weeks is preferable, as bile bacteria may contribute substantially to the risk of opportunistic pathogenic infection post-PD. In this study, the majority of patients had post-PBD drainage durations clustered within 3 weeks (86.3%). Our findings demonstrated that when PBD drainage lasted less than 10 days, there was a marked increase in both mortality and the rate of severe complications following PD. While there was no statistically significant difference between groups in specific major postoperative complications of PD, we observed that the incidence of grades B and C hemorrhage as well as the postoperative transfusion rate were significantly elevated in the cohort with PTCD drainage durations less than 10 days. Evidence supporting PBD’s ability to reduce PPH after PD has been substantiated by numerous studies [28]. However, insufficient drainage time may result in incomplete restoration of hepatic function, including impaired coagulation and inadequate vitamin K repletion. Moreover, our data also revealed increased rates of abdominal and pulmonary infections in patients with PTCD durations less than 10 days after PD. This finding reinforces that, despite the lower probability of PTCD-induced intraoperative biliary tract infection compared to endoscopic retrograde cholangiopancreatography or ENBD [30], the use of prophylactic antibiotics during the PBD drainage period and perioperative care requires careful consideration. Therefore, although we advocate for prompt surgical intervention following PBD, a minimum drainage period of 10 days, along with appropriate supportive treatment, is necessary to achieve optimal preoperative status.
As a single-center retrospective study, this research demonstrates reliability in data collection and grading evaluation. Nevertheless, some degree of selection bias between the PBD group and the DS group cannot be avoided. Furthermore, the study is subject to limitations, notably that only PTCD was used as the PBD type, and the sample size became insufficient after subgrouping. Moving forward, we intend to conduct a multicenter prospective study, building upon these analytical findings, to better optimize perioperative management and treatment strategies for patients with obstructive jaundice undergoing PD.
In conclusion, routine PBD is not advised for patients with mild to moderate obstructive jaundice scheduled for pancreaticoduodenectomy. However, for patients with TB > 256 µmol/L accompanied by systemic functional impairment requiring intervention, PBD should be considered. Ensuring suitable care and supportive therapy post-PBD, with continuous drainage maintained for at least 10 days before surgery, contributes to the reduction of severe postoperative complications and mortality.
ACKNOWLEDGEMENTS
We express our gratitude to all patients and to the multidisciplinary hepato-biliary-pancreatic cancer team members at this institution for their guidance. It is our hope that these findings will provide future clinical benefit.
Funding Statement
FUNDING This work was supported by the Science and technology project of Nantong City (JC2024080).
Footnotes
CONFLICT OF INTEREST
No potential conflict of interest relevant to this article was reported.
AUTHOR CONTRIBUTIONS
Conceptualization: YL. Data curation: CW, XF. Methodology: PW. Visualization: JW, YH. Writing - original draft: QG, DW, QC. Writing - review & editing: YL.
REFERENCES
- 1.Chen SC, Shyr YM, Wang SE. Longterm survival after pancreaticoduodenectomy for periampullary adenocarcinomas. HPB (Oxford) 2013;15:951–957. doi: 10.1111/hpb.12071. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wang D, Lin H, Guan C, Zhang X, Li P, Xin C, et al. Impact of preoperative biliary drainage on postoperative complications and prognosis after pancreaticoduodenectomy: a single-center retrospective cohort study. Front Oncol. 2022;12:1037671. doi: 10.3389/fonc.2022.1037671. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Chen Y, Ou G, Lian G, Luo H, Huang K, Huang Y. Effect of preoperative biliary drainage on complications following pancreatoduodenectomy: a meta-analysis. Medicine (Baltimore) 2015;94:e1199. doi: 10.1097/MD.0000000000001199. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Pavlidis ET, Pavlidis TE. Pathophysiological consequences of obstructive jaundice and perioperative management. Hepatobiliary Pancreat Dis Int. 2018;17:17–21. doi: 10.1016/j.hbpd.2018.01.008. [DOI] [PubMed] [Google Scholar]
- 5.Ishiwatari H, Sato J, Sakamoto H, Doi T, Ono H. Current status of preoperative endoscopic biliary drainage for distal and hilar biliary obstruction. Dig Endosc. 2024;36:969–980. doi: 10.1111/den.14786. [DOI] [PubMed] [Google Scholar]
- 6.van der Gaag NA, Rauws EA, van Eijck CH, Bruno MJ, van der Harst E, Kubben FJ, et al. Preoperative biliary drainage for cancer of the head of the pancreas. N Engl J Med. 2010;362:129–137. doi: 10.1056/NEJMoa0903230. [DOI] [PubMed] [Google Scholar]
- 7.Endo Y, Noda H, Watanabe F, Kakizawa N, Fukui T, Kato T, et al. Bridge of preoperative biliary drainage is a useful management for patients undergoing pancreaticoduodenectomy. Pancreatology. 2019;19:775–780. doi: 10.1016/j.pan.2019.06.013. [DOI] [PubMed] [Google Scholar]
- 8.Shen Z, Zhang J, Zhao S, Zhou Y, Wang W, Shen B. Preoperative biliary drainage of severely obstructive jaundiced patients decreases overall postoperative complications after pancreaticoduodenectomy: a retrospective and propensity score-matched analysis. Pancreatology. 2020;20:529–536. doi: 10.1016/j.pan.2020.02.002. [DOI] [PubMed] [Google Scholar]
- 9.Clavien PA, Barkun J, de Oliveira ML, Vauthey JN, Dindo D, Schulick RD, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg. 2009;250:187–196. doi: 10.1097/SLA.0b013e3181b13ca2. [DOI] [PubMed] [Google Scholar]
- 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;240:205–213. doi: 10.1097/01.sla.0000133083.54934.ae. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Bassi C, Marchegiani G, Dervenis C, Sarr M, Abu Hilal M, Adham M, et al. The 2016 update of the International Study Group (ISGPS) definition and grading of postoperative pancreatic fistula: 11 years after. Surgery. 2017;161:584–591. doi: 10.1016/j.surg.2016.11.014. [DOI] [PubMed] [Google Scholar]
- 12.Wente MN, Veit JA, Bassi C, Dervenis C, Fingerhut A, Gouma DJ, et al. Postpancreatectomy hemorrhage (PPH): an International Study Group of Pancreatic Surgery (ISGPS) definition. Surgery. 2007;142:20–25. doi: 10.1016/j.surg.2007.02.001. [DOI] [PubMed] [Google Scholar]
- 13.Koch M, Garden OJ, Padbury R, Rahbari NN, Adam R, Capussotti L, et al. Bile leakage after hepatobiliary and pancreatic surgery: a definition and grading of severity by the International Study Group of Liver Surgery. Surgery. 2011;149:680–688. doi: 10.1016/j.surg.2010.12.002. [DOI] [PubMed] [Google Scholar]
- 14.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 the International Study Group of Pancreatic Surgery (ISGPS) Surgery. 2007;142:761–768. doi: 10.1016/j.surg.2007.05.005. [DOI] [PubMed] [Google Scholar]
- 15.Pattarapuntakul T, Charoenrit T, Netinatsunton N, Yaowmaneerat T, Pitakteerabundit T, Ovartlarnporn B, et al. Postoperative outcomes of resectable periampullary cancer accompanied by obstructive jaundice with and without preoperative endoscopic biliary drainage. Front Oncol. 2022;12:1040508. doi: 10.3389/fonc.2022.1040508. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Shen Z, Xu Z, Wang W, Xu W, Zhou Y, Lu X, et al. A novel nomogram for predicting the risk of major complications after pancreaticoduodenectomy in patients with obstructive jaundice. Clin Chim Acta. 2021;517:162–170. doi: 10.1016/j.cca.2021.02.018. [DOI] [PubMed] [Google Scholar]
- 17.Gong S, Song S, Cheng Q, Huang Y, Tian H, Jing W, et al. Efficacy and safety of preoperative biliary drainage in patients undergoing pancreaticoduodenectomy: an updated systematic review and meta-analysis. Expert Rev Gastroenterol Hepatol. 2021;15:1411–1426. doi: 10.1080/17474124.2021.2013805. [DOI] [PubMed] [Google Scholar]
- 18.Bineshfar N, Malekpour Alamdari N, Rostami T, Mirahmadi A, Zeinalpour A. The effect of preoperative biliary drainage on postoperative complications of pancreaticoduodenectomy: a triple center retrospective study. BMC Surg. 2022;22:399. doi: 10.1186/s12893-022-01853-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Johansen K, Khodakaram K, Lundgren L, Hasselgren K, Gasslander T, Sandström P, et al. Preoperative biliary drainage does not independently reduce survival after pancreaticoduodenectomy among patients with pancreatic ductal adenocarcinoma: a national registry study. Ann Surg Open. 2021;2:e090. doi: 10.1097/AS9.0000000000000090. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Zhang B, Lang Z, Zhu K, Luo W, Zhao Z, Zhang Z, et al. Whether preoperative biliary drainage leads to better patient outcomes of pancreaticoduodenectomy: a meta-analysis and systematic review. BMC Gastroenterol. 2025;25:161. doi: 10.1186/s12876-025-03761-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Olecki EJ, Swinarska J, Perez Holguin RA, Stahl KA, Wong WG, Peng JS, et al. Is preoperative biliary stenting associated with increased rate of postoperative complications for patients undergoing pancreatoduodenectomy? A review of national surgical quality improvement program data. HPB (Oxford) 2022;24:1501–1510. doi: 10.1016/j.hpb.2022.01.006. [DOI] [PubMed] [Google Scholar]
- 22.Sun P, Zhong Y, Hu Y, Diwas S, Wu J, Zou R, et al. Preoperative biliary drainage for patients with malignant obstructive jaundice: an update on the systematic review and model-based Bayesian network meta-analysis. Int J Surg. 2025;111:5475–5484. doi: 10.1097/JS9.0000000000002559. [DOI] [PubMed] [Google Scholar]
- 23.Zhao XQ, Dong JH, Jiang K, Huang XQ, Zhang WZ. Comparison of percutaneous transhepatic biliary drainage and endoscopic biliary drainage in the management of malignant biliary tract obstruction: a meta-analysis. Dig Endosc. 2015;27:137–145. doi: 10.1111/den.12320. [DOI] [PubMed] [Google Scholar]
- 24.Dorcaratto D, Hogan NM, Muñoz E, Garcés M, Limongelli P, Sabater L, et al. Is percutaneous transhepatic biliary drainage better than endoscopic drainage in the management of jaundiced patients awaiting pancreaticoduodenectomy? A systematic review and meta-analysis. J Vasc Interv Radiol. 2018;29:676–687. doi: 10.1016/j.jvir.2017.12.027. [DOI] [PubMed] [Google Scholar]
- 25.Yamamoto Y, Sugiura T, Esaki M, Takahashi Y, Arita J, Hashimoto M, et al. Impact of biliary drainage method before pancreaticoduodenectomy on short- and long-term outcomes in patients with periampullary carcinoma and obstructive jaundice: a multicenter retrospective analysis. Surgery. 2024;176:616–625. doi: 10.1016/j.surg.2024.05.011. [DOI] [PubMed] [Google Scholar]
- 26.Wang W, Zhang C, Li B, Yuan GY, Zeng ZW. Clinical evaluation of endoscopic biliary stenting in treatment of malignant obstructive jaundice. World J Gastrointest Surg. 2025;17:97596. doi: 10.4240/wjgs.v17.i1.97596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Maatouk M, Kbir GH, Ben Dhaou A, Nouira M, Chamekh A, Daldoul S, et al. Pancreatic surgery after preoperative biliary drainage in periampullary cancers: does timing matter? A systematic review and meta-analysis. HPB (Oxford) 2025;27:10–20. doi: 10.1016/j.hpb.2024.10.001. [DOI] [PubMed] [Google Scholar]
- 28.Cao H, Li T, Li Z, Zhao B, Liu Z, Wang W. Goal-oriented preoperative biliary drainage is more precise and conducive to seize the opportunity for pancreaticoduodenectomy. World J Surg Oncol. 2024;22:331. doi: 10.1186/s12957-024-03615-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Zhu L, Yang Y, Cheng H, Cai Z, Tang N, Mao L, et al. The role of preoperative biliary drainage on postoperative outcome after pancreaticoduodenectomy in patients with obstructive jaundice. Gland Surg. 2023;12:593–608. doi: 10.21037/gs-22-648. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Wu MJ, Chan YY, Chen MY, Hung YL, Kou HW, Tsai CY, et al. The clinical impact of different types of preoperative biliary intervention on postoperative biliary tract infection of patients undergoing pancreaticoduodenectomy. J Clin Med. 2024;13:4150. doi: 10.3390/jcm13144150. [DOI] [PMC free article] [PubMed] [Google Scholar]