Abstract
Background/Aim: Preoperative biliary drainage (PBD) is often performed for jaundiced patients. However, the optimal duration of PBD remains unknown. The aim of this study was to investigate whether duration of PBD influenced the prognosis of patients after pancreaticoduodenectomy (PD) for pancreatic head cancer. Patients and Methods: Twenty-five patients who underwent PD for pancreatic head cancer with obstructive jaundice between 2007 and 2013 were included. Tumor and host factors were analyzed to evaluate their potential prognostic effects and patients' characteristics between the two groups according to the duration of PBD were analyzed. Results: In multivariate analysis, overall survival, duration of PBD ≥21 days and tumor-node-metastasis (TNM) stage III or IV were significant predictors. Duration of PBD ≥21 days was positively correlated with higher level of serum C-reactive protein (CRP), modified Glasgow prognostic score (mGPS) and neoadjuvant therapy. Conclusion: Duration of PBD is an independent prognostic factor after PD for pancreatic head cancer with obstructive jaundice.
- Pancreatic head cancer
- obstructive jaundice
- preoperative biliary drainage
- duration of preoperative biliary drainage
Obstructive jaundice is the most common symptom in patients with periampullary cancer, including pancreatic head cancer. For patients with a resectable tumor who have no radiological evidence of metastasis, surgical resection is the only option for cure (1, 2). Preoperative biliary drainage (PBD) remains controversial in the evaluation and management of patients prior to pancreaticoduodenectomy (PD) (3). A large scale of randomized control trials, reported by van des Gaag et al. demonstrated that routine PBD in patients undergoing surgery for pancreatic head cancer should not be undertaken because of the increased of rate of complications, in particular, preoperative complications related to biliary drainage, and that surgery should be realized as soon as possible. However, PBD is often performed for jaundiced patients with periampullary cancer, not only because a few weeks are often required before surgery, but also because PBD permits patients to undergo neoadjuvant therapy as it relieves symptomatic jaundice and factors that are significant concerns in patients. Therefore, it is important to evaluate the optimal duration of PBD. On the other hand, to our knowledge, the prognostic value of duration of PBD has not been previously examined.
The present study aimed to investigate whether duration of PBD influenced the prognosis of patients after surgery for pancreatic head cancer with obstructive jaundice.
Materials and Methods
A series of 33 consecutive patients underwent resection for pancreatic head tumor between January 2007 and December 2013 at the Kawaguchi Municipal Medical Center (KMMC). Among the 33 patients, 8 patients were excluded; 5 patients due to absence of obstructive jaundice and 3 who died of other diseases. The remaining 25 patients who underwent preoperative biliary drainage were retrospectively reviewed. This study was approved by KMMC's research ethics committee.
Among our multidisciplinary treatment team, principal indications for PBD included right upper quadrant pain, pruritus, acute cholangitis and anticipation of several-week delay in preparation for curative resection because of medical comorbidities. In all patients with obstructive jaundice, endoscopic drainage was attempted. Patients with inability to cannulate the common bile duct underwent percutaneous biliary drainage.
The criteria for resectability used have been outlined elsewhere (1). The surgical procedures were PD, pylorus-preserving PD (PpPD) and subtotal stomach preserving PD (SSPPD) for tumor located in the pancreatic head or uncus. Superior mesenteric or portal vein resection was performed for suspected tumor infiltration. The operative specimen underwent standard histopathological evaluation by a dedicated and experienced pathologist. The tumor staging was based on tumor-node-metastasis (TNM) stage classified by the Union for International Cancer Control (UICC). UICC's residual tumor classification was used to define the radicality of resection. During statistical analysis, R1 and R2 resections were evaluated in the same sub-group. Patients were followed at regular intervals after discharge by blood tests or imaging studies every 6 months.
The original Glasgow prognostic score (GPS) was constructed as previously described (4). Patients with both elevated serum C-reactive protein (CRP) (10 mg/l) and hypoalbuminemia (35 g/l) were assigned a score of 2. Patients in whom only one abnormality was present were allocated a score of 1 and 0 if neither abnormality was present. Modification of the score has resulted in an elevated serum CRP being assigned a modified Glasgow prognostic score (mGPS) of 1 or 2 depending on the absence or presence of hypoalbuminemia (5).
The primary end-point was to evaluate whether duration of PBD influenced survival in univariate and multivariate models. The secondary end-point was to analyze patients' characteristics between two groups (shorter duration of PBD or longer duration of PBD) using the median value as the cut-off.
Statistics. Statistical analyses were performed using the Statistical Package for Social Sciences (SPSS 23.0 for Windows; SPSS. Inc., Chicago, IL, USA). For statistical analysis, variable groupings were performed by means of standard laboratory parameters thresholds. Continuous data were expressed as median and range and compared by the Mann-Whitney U-test between two groups. Categorical data were compared by Pearson's chi-square test. Univariate analysis was performed by the Kaplan-Meier method (log-rank test) and differences were evaluated using the log-rank test. Multivariate analysis, including all variables with p-values of <0.10 in univariate analysis, was performed by a stepwise backward procedure until all variables remaining in the model were significant. A p-value of <0.05 was considered statistically significant in all analyses.
Results
Clinical patients' characteristics. The background characteristics of the 25 patients are summarized in Table I. There were 14 men and 11 women. The median age was 67 (range=38-87) years. The median follow-up time of surviving patients was 36.3 (range=18.8-85.7) months. Biliary drainage was accomplished endoscopically in 23 of 25 patients (92%), while 2 other patients required percutaneous procedures. The median duration of biliary drainage was 20 (range=8-168) days. Three patients underwent neoadjuvant therapy; one patient with suspected solitary lung metastasis received gemcitabine and the others with tumor encasement of the portomesenteric vein took gemcitabine alone or combined therapy with gemcitabine and S-1, respectively. One of the patients with portal invasion died 26.3 months after surgery; the others survived without recurrence for 19.7 months or 33.2 months after surgery, respectively. Disease-free margins (R0) were achieved in 16 patients (64%).
Variables associated with overall survival in univariate and multivariate analyses. Table II demonstrates the relationship between the clinicopathological variables and overall survival after surgery for pancreatic head cancer. In univariate analysis, overall survival was poor in patients with duration of PBD ≥21 days (p=0.0885), UICC TNM Stage III or IV (p=0.005), lymphantic invasion (p=0.0334) and perineural invasion (p=0.0339). In multivariate analysis, duration of PBD ≥21 days (p=0.0485), UICC TNM Stage III or IV (p=0.0064) were independent and significant predictors of poor overall survival.
Univariate analysis of patients' characteristics in relation to duration of PBD. Table III demonstrates the relationship between clinicopathological variables and duration of PBD. In univariate analysis, preoperative serum CRP (p=0.0148), neoadjuvant therapy (p=0.0373) and mGPS (p=0.0048) positively correlated with duration of PBD.
Discussion
The findings of the present study indicate that the independent prognostic factors after PD for pancreatic head cancer with obstructive jaundice are longer duration of PBD (≥21 days) and advanced UICC TNM Sage (Stage III or IV), while longer duration of PBD was significantly associated with higher scores of mGPS.
The optimal duration of biliary drainage before surgery has not been established. Experimental and clinical studies have suggested that a period of at least 4 to 6 weeks is needed for the restoration of normal major synthetic and clearance functions of the liver, as well as intestinal mucosal barrier functions (6-8). A short period of drainage may not lead to full recovery of the metabolic abnormalities associated with obstructive jaundice. In some studies, relating to the effect of PBD on postoperative complications after PD, median duration of PBD was 4 to 6 weeks (3, 9-11). Recently, it has been acknowledged that PBD may lead to an increased rate of preoperative complications (11) and should not be performed routinely (12). However, no study, including the aforementioned ones, has ever noted a relation between duration of PBD and prognosis. Interestingly, Sanjeevi et al. reported impact of delay between imaging and treatment in patients with potentially curable pancreatic cancer (13). Pancreatectomy within 32 days of diagnostic imaging reduced the risk of tumor progression to unresectable disease by half as compared with a longer waiting time, whereas the risk of progression to unresectable disease was negligible for surgery scheduled within 22 days of diagnostic imaging (13). The cut-off value of the imaging-to-resection/reassessment-interval in their report was 22 days, which was very similar to the cut-off value of duration of PBD of the current study (21 days). Duration of PBD might be associated with prognosis after pancreatectomy.
In the present study, longer duration of PBD was significantly associated with higher scores of mGPS. There are publications reporting that systemic inflammatory response, such as serum CRP or mGPS, is an independent prognostic factor in patients undergoing resection for pancreatic cancer (14-16). In the present study, preoperative cholangitis was higher in the group of longer duration of PBD than their shorter counterpart, although the difference was not significant (18% versus 7%, respectively; p=0.3992). The relation of longer duration of PBD to higher scores of mGPS might be not only due to acute inflammation, such as preoperative cholangitis, but also due to chronic cancer-related inflammation, which causes suppression of anti-tumor immunity through recruitment of regulatory T cells and activation of chemokines encouraging tumor growth and metastasis (17).
There are several limitations in this study. First, this is a retrospective study. The duration of PBD was not decided before PD, hence, the potential of bias and confounding factors exist. Second, the number of patients who underwent PD was small so that the survival rate after PD might be underestimated. To confirm the influence of duration of PBD on survival after PD, more case numbers or prospective studies are required.
In conclusion, prognosis after PD for pancreatic head cancer with obstructive jaundice is poor in cases with duration of PBD ≥21 days. PD for pancreatic head cancer with obstructive jaundice should be followed as soon as possible in terms of not only tumor progression but also from the viewpoints of systemic inflammatory response.
- Received March 24, 2017.
- Revision received April 12, 2017.
- Accepted April 14, 2017.
- Copyright© 2017, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved